Slashdot Mirror


Doom Creator Says Direct3D Is Now Better Than OpenGL

arcticstoat writes "First-person shooter godfather and OpenGL stickler John Carmack has revealed that he now prefers Direct3D to OpenGL, saying that 'inertia' is the main reason why id Software has stuck by the cross-platform 3D graphics API for years. In a recent interview, the co-founder of id Software said, 'I actually think that Direct3D is a rather better API today.' He added, 'Microsoft had the courage to continue making significant incompatible changes to improve the API, while OpenGL has been held back by compatibility concerns. Direct3D handles multi-threading better, and newer versions manage state better.'"

405 comments

  1. DirectX by devxo · · Score: 1, Interesting

    It's been true for almost 10 years. In fact Microsoft's support for DirectX has always been better than what OpenGL had. Microsoft made it easy to use with all their programming tools and languages and had a great documentation. The API was always cleaner too. There were tons of books written for DirectX. This is the area Microsoft handles extremely well - their Visual Studio development environment is the best IDE on the market and they create great tools for developers. Their mobile development tools kick Apple's and Google's (C#, Visual Studio and Silverlight against Java...).

    It would be nice to see open source community wake up and start developing a competitor, as just now Microsoft is the driving force that innovates new technologies for PC and Xbox360 graphics and gaming. But for once it looks like the fact they're the only one doing so isn't slowing them down - they do a good job.

    1. Re:DirectX by C_me_glow · · Score: 2

      it would be nice. but open source isn't about nice.

    2. Re:DirectX by Anonymous Coward · · Score: 0, Troll

      No, it's a religion.

    3. Re:DirectX by Anonymous Coward · · Score: 0, Troll

      ...their Visual Studio development environment is the best IDE on the market...

      +5, Funny?

    4. Re:DirectX by Anonymous Coward · · Score: 5, Insightful

      it would be nice. but open source isn't about nice.

      And that is why Open Source doesn't win. Be nice and user friendly, and you are able to play better with others.

    5. Re:DirectX by tgrigsby · · Score: 1, Offtopic

      ...their Visual Studio development environment is the best IDE on the market...

      +5, Funny?

      Well, +5 Fanboi isn't one of the options Slashdot gives you, so yeah, +5 Funny.

      --
      *** *** You're just jealous 'cause the voices talk to me... ***
    6. Re:DirectX by Anonymous Coward · · Score: 0

      In fact Microsoft's support for DirectX has always been better than what OpenGL had

      "Better" depends on your purpose. If you want to use Direct3D on a non-Windows platform, Microsoft's support has always been much, much worse than what OpenGL has.

      It would be nice to see open source community wake up and start developing a competitor

      Direct3D doesn't compete on open source platforms at all. Microsoft will need to first START COMPETING before the open source community would be spurred into action. I'd love to see that happen though.

    7. Re:DirectX by sexconker · · Score: 5, Funny

      It's been true for almost 10 years. In fact Microsoft's support for DirectX has always been better than what OpenGL had. Microsoft made it easy to use with all their programming tools and languages and had a great documentation. The API was always cleaner too. There were tons of books written for DirectX. This is the area Microsoft handles extremely well - their Visual Studio development environment is the best IDE on the market and they create great tools for developers. Their mobile development tools kick Apple's and Google's (C#, Visual Studio and Silverlight against Java...).

      It would be nice to see open source community wake up and start developing a competitor, as just now Microsoft is the driving force that innovates new technologies for PC and Xbox360 graphics and gaming. But for once it looks like the fact they're the only one doing so isn't slowing them down - they do a good job.

      We know who you are!
      We have you surrounded!

      Step away from the chair and come out with your hands up.

    8. Re:DirectX by MrEricSir · · Score: 4, Insightful

      Say what you will about the confusing configuration dialogs and lack of build config options. Visual Studio + Visual Assist is (so far) untouched in terms of features and stability.

      (And I'm saying this as a full time developer of Gnome apps...)

      --
      There's no -1 for "I don't get it."
    9. Re:DirectX by Anonymous Coward · · Score: 0

      It would be nice to see open source community wake up and start developing a competitor,

      Oh, that's me, I use OpenGL.
      I am just going to waltz into the nvidia headquarters and ask what they're doing and how I can develop a fork of OpenGL based on that.

      Then I'll make a super-popular operating system and bundle that new API with it, throw in some marketing, etc.

      Imagine the things I could do if I woke up AND had breakfast!

    10. Re:DirectX by mangu · · Score: 1

      Say what you will about the confusing configuration dialogs and lack of build config options. Visual Studio + Visual Assist is (so far) untouched in terms of features and stability.

      (And I'm saying this as a full time developer of Gnome apps...)

      Had you been a Kde developer you'd know kdevelop, which is truly the best development environment of them all.

    11. Re:DirectX by errandum · · Score: 0, Flamebait

      Visual Studio development environment is the best IDE on the market.

      Just because you've only actively used that one, doesn't make it "the best IDE on the market". I, for one, have used quite a few in 3 different platforms and I have no patience for visual studio and I cringe every time I have to do anything ".NET".

    12. Re:DirectX by Nadaka · · Score: 1

      Visual Studio is pretty damn good. May not be the best, but it is certainly way up there. And I can easily see a lot of people preferring it.

    13. Re:DirectX by Cowmonaut · · Score: 4, Insightful

      And how many games are made for Apple compared to Windows?

    14. Re:DirectX by calmofthestorm · · Score: 2, Funny

      OP spelled vim/emacs wrong.

      --
      93rd rule of Slashdot: No matter how obvious my sarcasm is, my comment will be taken seriously by someone.
    15. Re:DirectX by Anonymous Coward · · Score: 0

      And now you can port your windows/x360 games to windows phone too. Just give the phone another generation or two of hardware upgrade, and xbox indie arcade will be on the phone.

    16. Re:DirectX by Anonymous Coward · · Score: 0

      Enlighten us: what do you use that is so much better than Visual Studio? And what is cringe-inducing in .NET?

    17. Re:DirectX by Eponymous+Coward · · Score: 1, Informative

      Visual Studio is ok, but it's very overpriced and poorly supported by Microsoft.

      My favorite IDE is IDEA from Jetbrains.

    18. Re:DirectX by VGPowerlord · · Score: 1

      In fact Microsoft's support for DirectX has always been better than what OpenGL had

      "Better" depends on your purpose. If you want to use Direct3D on a non-Windows platform, Microsoft's support has always been much, much worse than what OpenGL has.

      Except on the Xbox 360.

      --
      GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
    19. Re:DirectX by MrEricSir · · Score: 2, Funny

      Interesting how praising a Microsoft product now gets you modded as "Flamebait."

      Slashdot, where the echochamber is too fragile to be disturbed by contrary opinions.

      --
      There's no -1 for "I don't get it."
    20. Re:DirectX by errandum · · Score: 1

      My first annoyance is visual studio. Too much bloat and hard to get to work the way you want it. Second, .NET is windows only. If you program for a smartphone or plan to port your applications to any other OS, you're screwed. But I know some people take it as a religion, so I don't won't go there. I just don't like the lack of freedom. for multi-platform development I use Eclipse if it is a big project. It works for C/C++, java, python and "android". I don't dislike XCode (way less bloat than Visual Studio), although apple seems set on destroying it by taking it in that direction. I've also used codeblocks in the past due to it's multi-platform consistency. If I'm working on a windows machine and just editing a few lines of code, I use Notepad++. If I'm on a mac, I use Textmate (this one is my favorite). And while I can see the merits of Visual Studio, I don't like it and would much rather not use it.

    21. Re:DirectX by Half-pint+HAL · · Score: 1

      Does Visual Studio automatically fix the tab order for GUI elements yet? As a heavy keyboard user, I had my fill of amateur VS-coded interfaces a long, long time ago.

      --
      Got them moderator blues I blieve I walk out the do', With these mod-points I been gettin', I 'most never post no mo'
    22. Re:DirectX by sproketboy · · Score: 0

      I was seriously expecting a /s here. Visual Studio the best IDE? LOL. Visual Notepad compared to IntelliJ.

    23. Re:DirectX by Anonymous Coward · · Score: 0

      Looking at the ipod/ipad app store...i might say a lot more

    24. Re:DirectX by Tharsman · · Score: 1

      Visual Studio development environment is the best IDE on the market.

      Just because you've only actively used that one, doesn't make it "the best IDE on the market". I, for one, have used quite a few in 3 different platforms and I have no patience for visual studio and I cringe every time I have to do anything ".NET".

      I am a huge "Apple Fanboi". I started my non-educational development with CodeWarrior back in 1998. I worked with the horrendous IDE apple provided back in the pre-OSX days and also with RealBasic. I have worked with XCode. I have worked with Eclipse, and I have worked with Visual Studio. Visual Studio has, for me, been the best IDE in the world. Fast, responsive, the best auto-complete I have tried, and the debugger is just a bliss to work with.

      Your closing comment, though, makes it sound like your grudge is not with the IDE but with .NET. If you can't isolate one from the other then you can't make an objective call on the quality of any IDE, your view will always be biased by the language you are coding in.

    25. Re:DirectX by RightSaidFred99 · · Score: 0

      Nobody takes these people seriously. Visual Studio is arguably the best IDE there is, it doesn't matter how a bunch of clueless, angry dweebs mod someone for saying it.

    26. Re:DirectX by Danathar · · Score: 4, Insightful

      What lunacy...I guess Linux didn't go anywhere either cause it's open source...or Chrome.....or Firefox.....

    27. Re:DirectX by Anonymous Coward · · Score: 0

      Yes - for silverlight and wpf.
      Not sure about current version for winforms

    28. Re:DirectX by RightSaidFred99 · · Score: 1

      Yet more generalized arguments from the "it's 1998!" crew. It's not "hard to get to work the way you want it", and VS2010 does not feel bloated. Good luck porting that Android app to iPhone because you were smart enough to use Eclipse, btw. That'll work out real well for you.

    29. Re:DirectX by h4rr4r · · Score: 1

      Which is running a PowerPC build of windows. Meaning that his statement still stands.

    30. Re:DirectX by Anonymous Coward · · Score: 1

      8 years. D3D 8 was the first version really competitive with OpenGL, while D3D 9 pulled ahead.

    31. Re:DirectX by Anonymous Coward · · Score: 0

      I am sorry but what is better? There are a lot of good IDE's out there but I cannot think of one that is better or even close to as good. I am not trolling here but in all honesty what can really compare. I mean yes there is the price concern, but otherwise, what can you really compare it to?

    32. Re:DirectX by the+linux+geek · · Score: 2

      I like Visual Studio, even though I'm still not 100% sold on .NET (Java and Android developer by trade.) I'd say it's probably the best IDE out there. I've had nothing but bad luck with Eclipse, including a persistent issue where the menu bars disappear after an exit on Linux, and the only way to get them back is to clear out .workspace.

      I found Code::Blocks to be awful when I used it.

    33. Re:DirectX by FictionPimp · · Score: 1

      I am a fulltime apple user that sometimes has to do some .NET work. I don't mind VS. It is nice and get's the job done. Personally though I want my IDE to be as light and as out of my way as possible. I enjoy XCode, but really I do most of my work in textmate (although when I'm doing web development I tend to use Coda).

      I can't find any faults with VS except for it's just a little too heavy for my liking, but it's still a rock solid IDE.

    34. Re:DirectX by Eponymous+Coward · · Score: 2

      the debugger is just a bliss to work with

      Microsoft makes a fantastic debugger, but it isn't the one in Visual Studio. I'm talking about WinDbg. It's the best debugger I have used on any platform.

      My problem with Visual Studio isn't with the functionality it has (although I find it slow), it's what is missing. The biggest problem for me is the lack of refactoring support. There are plugins that add this, but they add $250 to the $700 price tag (for the professional version).

    35. Re:DirectX by kelemvor4 · · Score: 1

      There's no Direct X in Apple products either, but nobody says "And that is why Apple doesn't win."

      There's plenty of other reasons apple products don't win. No need to dig all the way down to the nitty gritty of graphics api's to have that argument.

    36. Re:DirectX by Chelloveck · · Score: 3, Insightful

      "Commercial will always triumph over open, because open is dumb." -- Dark Helmet

      Or something like that.

      --
      Chelloveck
      I give up on debugging. From now on, SIGSEGV is a feature.
    37. Re:DirectX by Anonymous Coward · · Score: 0

      chrome isn't really open source, not in the way you're thinking. the code is open, but only google employees can commit to the actual product.

      FF is a shining example of how OSS can be done right. linux is not, at least not if you're talking about general use.

    38. Re:DirectX by errandum · · Score: 1

      Did you even read what I wrote? Or maybe you have no idea what XCode is. Clearly you know what you're talking about!

    39. Re:DirectX by emj · · Score: 1

      Whatever IDE you use the code still needs to be written.

      It would be nice to know why you feel so strongly against Eclipse and everything else that is not Visual Studio. Looking at the people who get computer science master degrees I would say a very large precentage of these people do not use Visual Studio, I'm just using that as an example because that is what I know.

      I do with you agree that almost nothing feels bloated with a good machine bought in this day and age (except some badly written Java and .net apps).

    40. Re:DirectX by Tetsujin · · Score: 3, Insightful

      it would be nice. but open source isn't about nice.

      And that is why Open Source doesn't win.

      I feel like the whole idea that we had to "win" in the first place was a fallacy...

      --
      Bow-ties are cool.
    41. Re:DirectX by errandum · · Score: 1

      Code blocks appears date now, but it was actually... nice a few years ago. No headaches, things just worked and I like that. Never had any kind of trouble with Eclipse, so I won't comment on that. But when a piece of software requires 2.3GB of HD space for a simple Visual Basic installation (according to http://www.microsoft.com/visualstudio/en-us/products/2010-editions/visual-basic-express ), is there any question about bloat? The whole point here is "visual studio is the best". It isn't, not in my opinion and many others. Is it good? Maybe, I don't like it, but I can see how some might. But stating that something is "the best" should be more than personal preference or ignorance about alternatives.

    42. Re:DirectX by nautsch · · Score: 1

      Yeah because everybody can freely commit to the firefox codebase.

      And would you care to elaborate the your linux remark? I still think that this is the most successful open source project.

      --
      If you find a typo, you may keep it.
    43. Re:DirectX by EnglishTim · · Score: 1

      Seriously though; is there anything that compares with it?

    44. Re:DirectX by EnglishTim · · Score: 1

      Has it improved greatly over the last few years? Last time I used it (admittedly half a decade ago), it didn't come close to Visual Studio.

    45. Re:DirectX by Whatsmynickname · · Score: 2

      You say Visual Studio is too much bloat. OK. But then you cite Eclipse as a counter-example?!?!?!? Hell, for example, Eclipse takes 10x longer to just start vs. Visual Studio (for my machine), let alone compile... Just about every IDE I've seen in 2011 has bloat, IMHO.

    46. Re:DirectX by Anonymous Coward · · Score: 0

      Used is not the same as winning. But it could had got a lot further if OOP people would do more. Documentation for OOP pretty much is hard to find. So be nice and do it better.

    47. Re:DirectX by Wumpus · · Score: 1

      Yes, but I just had lunch so I won't describe it.

    48. Re:DirectX by bhcompy · · Score: 2

      You mean those converted Flash games? Yea... That's like saying all my Nesticle games were made for Windows. Escape Velocity was the best Apple exclusive game, but it's no longer exclusive.

    49. Re:DirectX by giuseppemag · · Score: 1

      Express editions? Dreamspark? MSDNAA?

      --
      My book: Friendly F#, fun with game development and XNA; my game: Galaxy Wars by VSTeam; my gamedev language: Casanova.
    50. Re:DirectX by meerling · · Score: 1

      Of course that's like comparing a new 2010 Corvette Grand Sport to a new 1932 Ford Coup. One is an old tech car even though it's still nice and shiny.
      The capabilities of phones and mp3 players (other than touchscreen and pocketablity), are far inferior to those of a modern desktop or laptop no matter how you slice it. They don't have the storage, the screen res, the graphics processing power, the raw cpu power, the memory, etc.
      Sure, they've got some great games, and they've got 3000 slightly different versions of Tetris and Bejeweled.

      I guess the short of it can be summed up like this: The istuff (phone/pod/pad) platform does have a number of games (even after you eliminate minor variations) it can in no significant way compare to the games that take advantage of even a small portion of the capabilities available to a platform such as a modern Mac or PC.

      Now please redirect your attention to the main subject of the thread, Direct3D vs OpenGL.

    51. Re:DirectX by Anonymous Coward · · Score: 0

      From a strictly user perspective, they both look pretty damn nice, to me. I guess I'm not overly discriminating when it comes to, say, video detail - the absolute cutting edge just doesn't matter, especially if the game isn't all that fun to begin with. What I do prefer about OpenGL, is that it can be used across the two platforms that I use, Linux and Windows, and because my "hot" system runs Linux, obviously (if the producer offers it) I'm going to purchase the Linux variant. In the heat of the game, for me, there's no difference between UT99, UT2003, and UT2004 - the skills drawn on to play the games are what makes it "good". The rendering engine doesn't really matter.

    52. Re:DirectX by boristhespider · · Score: 2

      He's clearly talking about desktop environments, given that this is a thread about DirectX which is not renowned for its heavy usage on servers, and his phrase "general use". On the desktop even the most hardcore Linux enthusiast would have to admit that the sub-1% penetration is pretty lacklustre, compared to OSX let alone Windows.

      Yes, there are plenty of things even in my flat that run Linux that aren't desktops/laptops; yes, Linux on the server is much bigger than Linux on the desktop is. But in penetrating the general market, Firefox has made much greater inroads against IE than Linux has against Windows. By that metric, Firefox is a lot more succesful than Linux. By other metrics you might say something different... and in assessing the general market you'd have to accept that Linux is still very much a minority interest. (Unlike perhaps parts of the GNU toolchain which with Linux and OSX combined is now hitting between 5% and 10% of the general market, depending on what figures you believe. They'd also be a candidate for "most succesful" if such a thing is at all meaningful.)

    53. Re:DirectX by Archangel+Michael · · Score: 1

      I see you beat me to the punchline. Nicely played

      --
      Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
    54. Re:DirectX by bhcompy · · Score: 1

      Of course most of us do it as a hobby and VSE is freeware. Those working in corporate environments don't worry about cost, anyways.

    55. Re:DirectX by kevinNCSU · · Score: 1

      I use eclipse about 90% of the time and Visual Studio only 10% but I'm 100% sure you shouldn't have immediately mentioned Eclipse as a better alternative if you're going to go with the "bloat" argument. I'm often tempted to put off critical OS security updates just because I know I'll have to wait for eclipse to open again after the restart.

    56. Re:directx by dave562 · · Score: 1

      carmack is only writing for directx because ...

      From the content of the article I was left with the impression that Carmack IS NOT writing DirectX code. At the end of the article he explicitly states that DirectX does not offer enough benefits to make it worth rewriting his development tools and code base to leverage it.

    57. Re:DirectX by dgatwood · · Score: 1

      Indeed. It is not always necessary to "win". Often it is merely necessary to not lose. (Alternatively, often it is merely necessary to not be a loser.)

      --

      Check out my sci-fi/humor trilogy at PatriotsBooks.

    58. Re:DirectX by errandum · · Score: 1

      I said eclipse was good because it works consistently crossplatform. I've never had any qualms with Eclipse after that (admittedly) longer boot time... And I ask you to search for Textmate, Ultraedit or even Notepad++. Very little bloat (or none whatsoever) and tons of functionality (and still used in 2011). And we're not talking minimalist interface like vi or something like that. These are full featured IDE's.

    59. Re:DirectX by meerling · · Score: 2, Interesting

      Don't forget this is slashdot where any praise of something MS does is likely to get you modded a troll, which itself is actually trolling.

      Like most companies, MS has done good and bad things, sometimes with the exact same action. To blindly label all Microsoft as evil and all things Linux as good is just illogical, unjust, and rather stupid. Unfortunately there's a lot of that around here.

      So here's a few opinions bound to start the flaming from the mindless:
          Microsoft has done many good things over the years.
          Linux isn't the ultimate OS, and has less than even chance of ever becoming it.
          Macs breakdown and have plenty of bugs and crashes.
          Too much choice bogs things down.
          Microsoft has done some rather heinous things over the years.
          A properly updated and configured Linux makes a really good desktop.
          Apples 'think like we want you to' design of products works well for many people.
          The lack of choices can be frustrating to anyone that's creative, or knows what they really want.
          The gasoline corps are laughing all the way to the bank with giant crocodiles tears for the current 'crisis'.

      Asbestos Undies upgraded with Nomex PJs, I'm ready for you :D

    60. Re:DirectX by Anonymous Coward · · Score: 0

      Linux, Chrome and Firefox are user friendly nowadays. So open source IS about being nice.

      I don't see your point.

    61. Re:DirectX by badboy_tw2002 · · Score: 1

      Brrrrrrr...thanks for playing try again. If you consider OS X and FreeBSD different OSs, then your statement is far from correct.

    62. Re:DirectX by westlake · · Score: 2

      What lunacy...I guess Linux didn't go anywhere either cause it's open source...or Chrome.....or Firefox.....

      The Mozilla Foundation lives and dies by the add click.Where would Firefox be without the port to Windows?

      Unrestricted net assets - Revenues and other support (2009)

      Royalties: $101,537,000
      Contributions: $50,000

      Note 2 - Summary of significant accounting practices

      (e) Receivables

      Receivables consist primarilly of amounts due from contracts with multiple search engine and information providers

      Mozilla Foundation and Subsidiaries

      As a desktop client OS, the traditional community-oriented Linux distribution may not be six feet under. But neither is it in the best of health:

      Net Applications (March)

      Linux 0.92%
      iOS 1.8%
      Android 0.5%
      Operating System Market Share

      Statcounter (March)

      Linux 1%
      Top 5 Operating Systems

      W3Schools (January)

      Win 7 31%
      Up 31% since January 2009
      Linux 5%
      Up 3% since March 2003
      OS PLatform Statistics

    63. Re:DirectX by migla · · Score: 1

      >>And that is why Open Source doesn't win.
      >I feel like the whole idea that we had to "win" in the first place was a fallacy...

      Would you say the same thing about freedom?

      --
      Some of my favourite people are from th US; Vonnegut, Chomsky, Bill Hicks.
    64. Re:DirectX by Coeurderoy · · Score: 3, Interesting

      And this is modded insightful ?
      Free Software is not a religion, it is a policy, and as strong political roots, altough these political roots are certainly not aligned to "bipartisan politics".

      The issue with open souce and OpenGL is that a large part of the implementation of OpenGl (at least the efficient implementation) does not depend on any open source activist/developper but on the good will of video card developpers.

      Now Microsoft has the "monopoly advantage" if they say now you need to cut off your left feet to implement Direct3D the videocard fabricant (Intel, Nvidia principally) will find somebody in charge of getting his or her feet cut off to keep the market.
      OpenGL has to reach a concensus...

    65. Re:DirectX by h4rr4r · · Score: 1

      Their kernels are significantly different but they are in the same family. The OSes are very different, osx does not even use X for example. Xbox windows and windows are far closer together.

      I think you know far less than you think you know.

    66. Re:DirectX by beelsebob · · Score: 1

      The XBox at least uses a version of the windows kernel. OS X does not use the BSD kernel.

    67. Re:DirectX by tibit · · Score: 1

      I think that Qt Creator's C++ parser works way better than Intellisense. I get consistently better results (on same projects) as far as code insight goes with Qt Creator than with Visual Studio 2008. It's quite often that Intellisense simply gives up for a couple minutes at a time. Maybe it works great for C# and such, but for C++ it's far from useable (for me).

      The only gripe I have with Qt Creator is that it still depends on the pile of excrement that gdb is :(

      --
      A successful API design takes a mixture of software design and pedagogy.
    68. Re:DirectX by Kevin+Stevens · · Score: 3, Informative

      Pretty telling that you couldn't actually respond with an answer.

      VS makes me want to code. It makes browsing code so much easier. Debugging is so nice that I will often fire up a debugger and step through code if I have a hard time figuring out what something is doing. I have been in several situations where my entire team couldn't find a multithreaded related crash using gdb, but I have never had a bug escape me when debugging through Visual Studio. You can actually control administer your machine a lot more efficiently through VS than you can by clicking through the control panel. Its easy on the eyes, though you don't get "hardcore" points w/ coworkers for a black background with neon text. In my opinion, coding is easier, bugs get found faster, code gets written faster. Compile error? click on the error, and it will take you to the offending line. Huge convenience. And note I am talking about raw C++ here- no managed code, .net, or any other stuff.

      Clearly there are drawbacks- Intellisense just stops working sometimes, for no apparent reason. The build system is different than make, and can be annoying at times, but its way easier than fighting with autotools, though CMAKE is a little nicer (though fortunately CMAKE can generate vcproj files). Integration w/ source control other than VSS is a pain- usually I just avoid that altogether and leave them unintegrated.

      I have tried alternatives- I used Emacs in college exclusively, until I got sick of a million meta commands and having to write programs just to configure my IDE the way I like it (a little exaggeration there, but not much). browsing between files on large projects was too cumbersome.

      I have tried Eclipse and its god awful view system, where god forbid I click the wrong place or hit the wrong button my whole screen just gets deranged. I find the Eclipse model very awkward, the plugins are buggy, and intellisense works even less frequently than VS. For Java, Eclipse is good.

      I tried QTCreator, which was alright, but seemed very focused around building QT apps. I just wanted to do general dev, and the UI was too Apple-ish for me- I don't need a shiny IDE, just one that is pleasant to look at and lets me get things done.

      I have not tried net beans for C++, mostly because Eclipse CDT has lead me to believe that java IDE's shoe-horned to C++ IDE's don't seem to work that well. I could give Code::Blocks another shot as well.

      So seriously, what do you recommend? And what specifically do you think your recommendation does better than Visual Studio?

    69. Re:DirectX by tibit · · Score: 1

      I don't think that the boot-up times are important at all. If you're spending a significant amount of time starting up your IDE, you're not using it much anyway, or you're doing something seriously wrong (tm). I'll have Qt Creator running for a week or two at a time, same for VS2008 running in a VM. I can't even tell you how long it takes it to start them up: I start them up so rarely.

      --
      A successful API design takes a mixture of software design and pedagogy.
    70. Re:DirectX by tibit · · Score: 1

      I've used VS2008 with C#, and there intellisense has been rock solid. With C++ it regularly hangs up, sometimes for a minute or more, and I cannot really depend on it -- it's very frustrating. I think that for C++ development, Qt Creator's code parser tools beat the crap out of VS2008 at least: I'm comparing them side-by-side all the time, on same code bases. I can't say how Eclipse is in C++ code insight department, but I've used it a bit for Java development and it felt rock solid and dependable -- just like VS2008 did for C#.

      --
      A successful API design takes a mixture of software design and pedagogy.
    71. Re:DirectX by badboy_tw2002 · · Score: 1

      Right, and Xbox uses the same windowing library? What exactly makes them the same OS? My contention is that the 360 is not running Windows, and that seems to be the point being made.

    72. Re:DirectX by Tetsujin · · Score: 1

      >>And that is why Open Source doesn't win.
      >I feel like the whole idea that we had to "win" in the first place was a fallacy...

      Would you say the same thing about freedom?

      I don't get it. Freedom is largely a separate issue from the question of whether free software would become a mainstream end-user OS.

      --
      Bow-ties are cool.
    73. Re:DirectX by Heretic2 · · Score: 1

      You apparently didn't use any Direct3D version prior to 7. 10-15 years ago, all DirectX was crap in particular Direct3D. I was playing Quake before glQuake came out. Maybe you don't remember that.

    74. Re:DirectX by Anonymous Coward · · Score: 0

      Yes, and you can often find it in the park after I walk my dog and forget to bring a plastic bag.

    75. Re:DirectX by h4rr4r · · Score: 1

      My contention is that it either close enough to windows, or not a real alternative anyway so it does not count as a non-windows platform.

    76. Re:DirectX by EnglishTim · · Score: 1

      *rimshot*

    77. Re:DirectX by Anonymous Coward · · Score: 0

      clearly you never used IntelliJ, although, you cam pretty much get the experience by installing resharper on visual studio

    78. Re:DirectX by bzipitidoo · · Score: 1

      I didn't find developing in MS's Visual Studio nearly as friendly. Last time I really used it was just before .NET came out. I see no reason why many of the bad things about it would have changed.

      Here are some of the problems with developing with MS. There's "help" on some library functions and applications that wasn't "just the facts", but instead was meaningless marketing hype. The last thing I want to spend time doing is wading through something that's supposed to be documentation but instead is a horrid, insult to our intelligence sales pitch that goes on and on about how some "powerful" feature is going to save time, effort and money, and "spark my imagination", etc. Now, normally when I run into a lack of documentation, the usual answer is "use the source, Luke". But this is MS, it's all proprietary, you can't just look at the source code. Then there were libraries that had been phased out, only you didn't learn that on the first bit of documentation you found. You found that out after you'd spent hours coding an application to use it, when you ran into a bug in the library and went back to the docs, or the web for answers. You'd have the option of fixing that bug, if only you had source, but no. Same problem if a library is missing a feature. Can't add it, no source. You could request that MS fix the bug or add the feature, but you will be ignored, or at best told to use some other library. The missing feature happened when I was using a video capture and encoding library, and could not find any way to set some camera settings available in the GUI. The solution I came up with was dirty, but it mostly worked: shove appropriate keypresses into the buffer, then call the routine that makes the user interface pop up. Other people had resorted to the same trick for that same problem. Then there is an additional kind of problem you have to consider. Could DRM be causing the bug you are trying to figure out?

      As for debugging, the old print statement is still king. Yeah, it's nice to have a debugger, with watches, breakpoints, inspection of variables, etc. That's great for some lame business logic or web app, or for quickly finding where a segfault happened, but not enough for complicated algorithms, where data must be displayed in a way that can be comprehended, and subtle errors will be easily missed. Or where the quantity of data is so great, and filtering out whichever parts are irrelevant is itself not trivial. When your display commands in the debugger get so complicated they ought to be checked in to a repository, you may as well just program that logic into the source code.

      --
      Intellectual Property is a monopolistic, selfish, and defective concept. It is "tyranny over the mind of man"
    79. Re:DirectX by Anonymous Coward · · Score: 0

      No, it's a religion.

      Free Software is a strategy, and if you think otherwise, then you have never been screwed by being dependent on proprietary software that isn't getting maintenance anymore. Lucky you, young'un.

    80. Re:DirectX by ZFox · · Score: 1
      You ought to try out JetBrain's IntelliJ for the Java side of things.

      Its easy on the eyes, though you don't get "hardcore" points w/ coworkers for a black background with neon text.

      And stop making fun of my black background and neon text that I'm staring at right now in Visual Studio (okay, okay, 1 window behind slashdot).

    81. Re:directx by Ruede · · Score: 1

      tbh i didnt read TFA but considering rage is made for directx, at least one person @ id has to write stuff for directx :)

    82. Re:DirectX by Anonymous Coward · · Score: 0

      Back in the day of Direct X 3 and 4, the Direct3D part was shit. Now, it is better.

      No one really cared about Microsoft's support of either OpenGL or Direct3D. Real support comes from the video card manufacturers. The OS doesn't matter.

    83. Re:DirectX by shutdown+-p+now · · Score: 1

      I think that Qt Creator's C++ parser works way better than Intellisense. I get consistently better results (on same projects) as far as code insight goes with Qt Creator than with Visual Studio 2008. It's quite often that Intellisense simply gives up for a couple minutes at a time.

      It's true. Both Qt Creator and VS2008 seem to use "guesswork" algorithm rather than a proper C++ parser - so as not to implement a full-fledged front end (templates are Turing complete, remember; and you need to handle them to provide fully accurate code completion and symbol navigation). On code using templates heavily (e.g. a good half of Boost), both usually fail - but Qt Creator seems to be more resilient in that it keeps working for things it can understand, whereas in VS the parser could just shut down. Then also, VS would cache C++ Intellisense information in those pesky .ncb files, which got corrupted every now and then - also effectively disabling the feature.

      In VS2010 things are different - MS purchased C++ front-end code from EDG (the same thing that drives Comeau C++ and Intel C++), and that is used for code completion now. The results are impressive: it provides working (and correct) code completion on insane TMP stuff like nested Boost lambdas.

    84. Re:DirectX by DurendalMac · · Score: 1

      And all of the above is true, more or less. What so many /.ers and geeks in general can't grasp is that there are people out there who have valid reasons for seeing things differently and wanting different things. It's sad, really.

    85. Re:DirectX by shutdown+-p+now · · Score: 1

      Last time I really used it was just before .NET came out.

      That would be VS6 - released in 1998; more than 12 years ago. A lot has changed since then.

      There's "help" on some library functions and applications that wasn't "just the facts", but instead was meaningless marketing hype

      There's certainly a bunch of articles like that on MSDN, usually tutorials or "how-to" things, but I've never seen any such in actual class/method/function documentation. When you press F1 when editor cursor is on a symbol, you invariably end up in a dry, technical article documenting what it does.

      Then there were libraries that had been phased out, only you didn't learn that on the first bit of documentation you found.

      Well yes, it's not like you're going to have "THIS IS DEPRECATED" in big red letters over every single documentation page for a particular library - just like, say, you don't have that for AWT javadocs. But if you go to the title page of a particular technology, and there is a better replacement available, it will tell you.

      As for debugging, the old print statement is still king. Yeah, it's nice to have a debugger, with watches, breakpoints, inspection of variables, etc. That's great for some lame business logic or web app, or for quickly finding where a segfault happened, but not enough for complicated algorithms, where data must be displayed in a way that can be comprehended, and subtle errors will be easily missed. Or where the quantity of data is so great, and filtering out whichever parts are irrelevant is itself not trivial. When your display commands in the debugger get so complicated they ought to be checked in to a repository, you may as well just program that logic into the source code.

      "Printf debugging" also fails in the scenarios you describe. Even more so when you have asynchrony and/or multithreading involved. VS, meanwhile, has some special debugging features to tackle that.

      As well, in a big project, adding or removing a print for debugging can mean a fairly lengthy recompile. Sometimes you have to do it, but using it as a main debugging technique is more often a waste of time on non-trivial projects.

    86. Re:DirectX by judeancodersfront · · Score: 2

      It's a nerd religion. I think it is pretty obvious that expecting FOSS everywhere is unrealistic but much like creationists they already picked their world view and will now defend it even as the evidence against it piles up. FOSS doesn't have enough solutions for paying the bills, especially for desktop software. Gosling came out and said this recently but the FOSS crowd still wants to follow St. Stallman around even though he clearly doesn't have all the answers.

    87. Re:DirectX by judeancodersfront · · Score: 1

      Linux is user friendly until it needs to be updated.

    88. Re:DirectX by tehdaemon · · Score: 1
      Dangit, I so wish that "almost nothing" and "badly written Java and .net apps" actually had something in common. It would make my job so much easier and more pleasant. For example we have one server that feels like it is running on a 486sx machine, with 4MB RAM. A poorly written and ported from Mac Java app is the culprit.

      Can't comment on the VS issue much, I haven't used it since '01.

      T

      --
      Laws are horrible moral guides, moral guides make even worse laws.
    89. Re:DirectX by Anonymous Coward · · Score: 0

      I'm assuming you made a serious reply. Linux hasn't gone very far in the home desktop market, still being behind MacOS and all, but it has admittedly done well in the server market, even with it's lack of user friendliness. Oh, and Chrome and Firefox are very user friendly, so these are bad examples of "open source that isn't nice" per the GPP. Don't just pick one part of an opponent's position and nitpick, thinking that you're being illuminating - try to understand the whole argument as stated in it's context, including what they're replying to - that user unfriendly open source doesn't win in the home market (which is what this article is about, since Direct3D is all about home gaming PC's - not servers). Peace.

    90. Re:DirectX by Anonymous Coward · · Score: 0

      What's Linux?

    91. Re:DirectX by bzipitidoo · · Score: 1

      As well, in a big project, adding or removing a print for debugging can mean a fairly lengthy recompile.

      Why? If you have done a proper job of writing the code, it should be a neatly separated bunch of modules. Doesn't take long to recompile 1 or 2 modules.

      And you trot out the old argument over multithreading. I wish people would cut that out. Everyone is so scared of parallel programming, talks as if everything we learned about serial execution is inapplicable, as if race conditions and deadlocks are diabolically mysteriously difficult to find. They don't have to be hard to find. Really! The main thing is not to write "parallel spaghetti code" so to speak. For instance, I always found the Dining Philosophers problem a little stupid, in that it is somewhat contrived. There are a number of ways to avoid deadlock in that problem, such as make one of the philosophers left-handed, or make the operation of picking up both forks atomic. Or just give each philosopher their own 2 forks so they don't have to share. Such bugs can be found with humble old print debugging. It is a little more difficult if the print statement changes the timing so that the problem no longer occurs, but that can happen in a debugger too.

      Yes, VS6 was what I used. I have used later versions, but not looked at the .NET part.

      But if you go to the title page of a particular technology, and there is a better replacement available, it will tell you.

      Not always. Documentation can quickly go stale. Easy to not have the latest if you don't update religiously.

      --
      Intellectual Property is a monopolistic, selfish, and defective concept. It is "tyranny over the mind of man"
    92. Re:DirectX by Anonymous Coward · · Score: 0

      Because .NET and Android were big back in '98. Get a clue troll.

    93. Re:DirectX by Lanteran · · Score: 1

      The 360 OS to my knowledge runs basically on a streamlined XP kernel. It was compiled for a totally different architecture; but you might notice how MS had the slight advantage of total access to D3D/Windows source code.

      --
      "People don't want to learn linux" hasn't been a valid excuse since '03.
    94. Re:DirectX by Daniel+Phillips · · Score: 1

      Linux hasn't gone very far in the home desktop market, still being behind MacOS and all

      Linux is actually making good progress in the desktop world. Look, w3schools shows Linux at 5%, an all time high and not that far behind Apple. I actually know normal people, not geeks, who have installed Linux entirely as their own idea. And of course, Linux rules the world in phones at the moment, and soon in tablets as well. It's only a short jump from tablets back to the desktop.

      --
      Have you got your LWN subscription yet?
    95. Re:DirectX by westlake · · Score: 2

      Now Microsoft has the "monopoly advantage" if they say now you need to cut off your left feet to implement Direct3D the videocard fabricant (Intel, Nvidia principally) will find somebody in charge of getting his or her feet cut off to keep the market.

      This is nonsense.

      The reality is that Microsoft works closely with the hardware and software developer so that everyone remains on the same page.

      Microsoft builds a consensus around what is possible and what is desirable - not only in video and sound, but in every aspect of PC gaming.

    96. Re:DirectX by westlake · · Score: 1

      Look, w3schools shows Linux at 5%, an all time high and not that far behind Apple. I actually know normal people, not geeks, who have installed Linux entirely as their own idea. And of course, Linux rules the world in phones at the moment

      The W3Schools stats have always been most generous to Linux.

      But a bare 3% growth in eight years is telling when compared to the vitality shown by its competitors.

      In North America, the iOS rules in mobile. Globally, it is Symbian. The Android is a strong competitor. But Android is unmistakably a Google production.

      Not a community oriented Linux distribution.

    97. Re:DirectX by oliverthered · · Score: 2

      I think if you look back a good few years (not long after doom 3 was released... just after Direct X 10 he said exactly the same thing)..

      Infact Janurary 2007 according to wikipedia.

      Some former critics of Direct3D acknowledge that now Direct3D is as good if not better than OpenGL in terms of capabilities and ease of use. In January 2007, John Carmack said that "DX9 is really quite a good API level. Even with the D3D side of things, where I know I have a long history of people thinking I'm antagonistic against it. Microsoft has done a very, very good job of sensibly evolving it at each step—they're not worried about breaking backwards compatibility—and it's a pretty clean API. I especially like the work I'm doing on the 360, and it's probably the best graphics API as far as a sensibly designed thing that I've worked with."

      --
      thank God the internet isn't a human right.
    98. Re:DirectX by Daniel+Phillips · · Score: 1

      A bare 3% growth in eight years is telling when compared to the vitality shown by its competitors.

      Rubbish. Expressed relative to a 2% base, reaching 5% market share represents a 150% expansion of the area of that wedge. And the whole pie grew. We all know what is going to happen when Microsoft's grip on OEMs finally slips. And it is beginning to slip now.

      Seems to me Ballmer and Gahaffi have quite a bit in common.

      --
      Have you got your LWN subscription yet?
    99. Re:DirectX by Mortlath · · Score: 1

      You haven't used Visual Studio in 13 years (VS 6.0 came out in 1998), but you feel like you know what problems it still has?

    100. Re:DirectX by tibit · · Score: 1

      So I guess that's testament to the quality of their own C++ front-end code, then? Or is it just that EDG's code is somehow better suited to on-the-fly parsing than their own code?

      --
      A successful API design takes a mixture of software design and pedagogy.
    101. Re:DirectX by shutdown+-p+now · · Score: 1

      VS2008 did not use the parser that is used by VC++ compiler. The reason why it wasn't done is because it's too tightly coupled to the rest of it - it's a code that evolved over more than a decade now, and it was never designed to be reusable originally.

      In contrast, EDG is reusable by design - they make a living licensing it to whoever needs to parse C++ for any reason.

    102. Re:DirectX by iampiti · · Score: 1

      Wow! Two microsoft-favoring first comments (http://apple.slashdot.org/comments.pl?sid=2033736&cid=35458946) on two stories pitting microsoft vs the competition. Do the pay you much?

    103. Re:DirectX by Joce640k · · Score: 1

      VC++ is the best, yes. The code produced by the compiler, particularly for floating point math, is an order of magnitude better than gcc. The IDE is pretty good.

      It's also free to download...

      --
      No sig today...
    104. Re:DirectX by DaVince21 · · Score: 1

      Only as far as DirectX is a religion.

      --
      I am not devoid of humor.
    105. Re:DirectX by Joce640k · · Score: 1

      I'm OpenGL fanboi number one but I'll be the first to admit that he has a point. I find myself leaning more and more towards Direct3D these days.

      OpenGL isn't going to die so long as it's the only API available on Macs and smartphones but it lost its driving force years ago and the development process is now painfully slow. A few committee meetings a year isn't going to produce a cutting edge graphics API while Microsoft is actively revving their updates.

      The other major problem OpenGL has is that there is a newer, cleaner version - it's called OpenGL ES, but for some reason they're refusing to allow people to use it on desktop machines. One API for desktops and another for handhelds will be the death of OpenGL. To stay relevant they need to deprecate classic OpenGL and push OpenGL ES as the primary API for all machines.

      --
      No sig today...
    106. Re:DirectX by Anonymous Coward · · Score: 0

      I beg your Micro something pardon but..Most people do not realize open source technology, like for example the files used in either Libreoffice or Open office are xml JAR zipped files to make add on access to them easy. The intent of open source is to allow all users, poor and rich alike, to take advantage of the work of the other users on the Worldwide Web so that everyone builds onto everyone else's work. This nice little openness idea makes open source a much broader package than proprietary software, and open source has a much brighter future than software arriving from the secret eat your financial lunch jungle.

      Also, please advise the absolute security of DirectX?

    107. Re:DirectX by fatphil · · Score: 1

      Much of the linux (explicitly I mean the kernel) development behind Android was done indeed not by the community, nor by google themselves, but by Nokia. Nokia have always had a policy of upstreaming first wherever possible, and that benefits all the competitors who use similar SoCs and peripherals. I believe that Google effectively even said "Thanks Nokia (suckers!)" back in the early days.

      --
      Also FatPhil on SoylentNews, id 863
    108. Re:DirectX by Anonymous Coward · · Score: 0

      Visual Studio development environment is the best IDE on the market

      You're kidding, right? I don't think you've used anything outside of Visual Studio for development.

    109. Re:DirectX by WorBlux · · Score: 1

      The very fact that it's available is in it's own way a win.

    110. Re:DirectX by Anonymous Coward · · Score: 0

      Linux is easier to update than Windows! On my system I just type 'yum -y update' and the whole system updates. If I so chose I could even do the equivalent in a gui, but using a terminal is quicker.

      In Windows it is go to WIndows Update, select updates, download and install, usually reboot and repeat because it never seems to fully update the system in one go. If I fully want to update the Windows system I then have to visit every website for the software I have installed and manually update them as well.

      Total time to update Linux - 5 mins.
      Total time to update Windows - anything between 5 mins and 3 hours!

    111. Re:DirectX by lordholm · · Score: 1

      Most of ES is available in the "normal" OpenGL. Further, even if D3D had a cleaner API (my impression is that it is C++ based), it fails in the following areas: not standardised, not portable, C++ API (making language bindings can be a big mess).

      Until MS fixes those three MAJOR issues, OpenGL will be lightyears better for any project, independent on how much nicer the API is.

      The only non-portable APIs that I can stomach are GUI APIs, as an application written with proper MVC can be easily ported anyway (i.e. keep the model part platform independent).

      --
      "Civis Europaeus sum!"
    112. Re:DirectX by Xest · · Score: 1

      "I have not tried net beans for C++, mostly because Eclipse CDT has lead me to believe that java IDE's shoe-horned to C++ IDE's don't seem to work that well. I could give Code::Blocks another shot as well."

      It's a shame, I too was massively dissapointed with Eclipse having long been a Visual Studio user (Borland C++ before that!). Having heard about Eclipse being the open source answer for Java, C++, PHP etc. to Visual Studio it seemed the obvious choice. Eclipse just isn't even in the same class as Visual Studio, Eclipse is slow, buggy, it's plugin system seems to regularly fuck up meaning if you really want to use multiple languages you're best off having multiple Eclipse installs. I don't like the workspace concept either, it's just stupid.

      I figured I'd give NetBeans ago, not expecting much because well, Eclipse is the one that got all the hype, and honestly I was pleasantly suprised. I actually really really like NetBeans, it's still not quite Visual Studio but it's much better than Eclipse, definitely gets a big thumbs up from me.

      I'll admit I haven't tried IntelliJ, I hear it get consistently positive nods from those who have used it. I hope to give it a go some day.

    113. Re:DirectX by Joce640k · · Score: 1

      Most of ES is available in the "normal" OpenGL.

      Only "most"...?

      Issues:
      a) When enough people use something it becomes a "de-facto" standard.
      b) Portable .... except for the fact that your desktop OpenGL code won't run on a handheld (and vice versa).
      c) It's a COM API - which is just as bad as a C API from a language-bindings point of view (eg. OpenGL).

      --
      No sig today...
    114. Re:DirectX by lordholm · · Score: 1

      I said "most" as a disclaimer, I have only written desktop GL code and not looked into ES.

      I just checked, and according to Khronos, OpenGL ES is a subset of desktop OpenGL, thus all ES code should run in a normal GL implementation.

      Regarding your A) reply, while a system may be a de-facto standard, it does not make it standardised. Standardisation normally means that there are multiple implementations from different vendors. This is the case with OpenGL, but I don't see any D3D libraries floating around for OS X, Linux, BSD or Solaris. D3D is no more a standard than Windows is. B) as clarified, ES is a subset of normal GL. C) COM APIs while making language bindings possible certainly does not enable portability to other platforms.

      Further, even if there where some minute non subset part of ES despite what Khronos writes, this will be minor and can be solved with a few #ifdefs. If I want to port a D3D app, I either have to re-implement everything in OpenGL, or implement the D3D API (using OpenGL). This is several tens of thousands of lines of code most likely compared to potentially a few hundred lines of #ifdefed code.

      So yeah... ES is a lot more portable.

      --
      "Civis Europaeus sum!"
    115. Re:DirectX by DeathFromSomewhere · · Score: 1

      Yup, everyone that disagrees with you must be paid off. You may wish to remove the tin foil hat before you post though.

      --
      -1 overrated isn't the same thing as "I disagree".
  2. OpenGL ES 2.0 breaks compatibility... by Anonymous Coward · · Score: 1

    ...and the API is actually quite small and simple.

    WebGL is awesome and simple. Moves all the "hard" work into creating shaders, which is where it belongs.

    1. Re:OpenGL ES 2.0 breaks compatibility... by BitZtream · · Score: 1

      As does OpenGL 3.0, completely throws out backwards compatibility.

      --
      Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
  3. "Doom creator"? by Anonymous Coward · · Score: 5, Insightful

    Is Slashdot not for nerds anymore? I never thought I'd see the day when John Carmack was described on Slashdot as "Doom creator".

    1. Re:"Doom creator"? by Samantha+Wright · · Score: 0

      If you'd like to go into an off-topic semantic rant about it, I would actually even charge Tom Hall with being the creator of the game (he wrote the original plot) and then our good man Romero with turning it into the tedious piece of crap that actually shipped. (Fun fact: if it had been possible to make Doom like Tom Hall intended, storytelling in FPSes would be five years ahead from where it is today.)

      But it's not exactly like we can say Bobby Prince wrote the engine, so someone has to take the credit, I guess.

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    2. Re:"Doom creator"? by Kenja · · Score: 2, Funny

      Better then "the jerk who brought us Daikatana".

      --

      "Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
    3. Re:"Doom creator"? by Anonymous Coward · · Score: 0

      D3D has gotta be easier than this hack from Quake:

      float Q_rsqrt( float number )
      {
              long i;
              float x2, y;
              const float threehalfs = 1.5F;
              x2 = number * 0.5F;
              y = number;
              i = * ( long * )
              i = 0x5f3759df - ( i >> 1 );
              y = * ( float * )
              y = y * ( threehalfs - ( x2 * y * y ) );
              return y;
      }

    4. Re:"Doom creator"? by Anonymous Coward · · Score: 3, Insightful

      The topic is about comparing one API to another.

      In that context, even if you were to go into a semantic rant about it, John Carmack is the creator of Doom.

    5. Re:"Doom creator"? by Howitzer86 · · Score: 1

      Better watch out or he'll 'make you his bitch'.

    6. Re:"Doom creator"? by Tanman · · Score: 4, Informative

      I know. He created commander keen, first, so that should have been the game they used.

    7. Re:"Doom creator"? by syousef · · Score: 1

      But it's not exactly like we can say Bobby Prince wrote the engine, so someone has to take the credit, I guess.

      I can think of someone who's looking for work and might be interested:

      http://www.youtube.com/watch?v=peW2p4GVt98

      --
      These posts express my own personal views, not those of my employer
    8. Re:"Doom creator"? by Anonymous Coward · · Score: 0

      That was John Romero, you insensitive clod.

    9. Re:"Doom creator"? by vgerclover · · Score: 2

      That's the Other John, Romero. Carmack is the great one.

    10. Re:"Doom creator"? by jones_supa · · Score: 1

      i = * ( long * )

      y = * ( float * )

      What's the idea behind these recasts?

    11. Re:"Doom creator"? by Opportunist · · Score: 1

      That's just one bit of magic rote. Nothing to learn here, just an (ab)use of an IEEE 754 format quirk.

      So "easy" and "hard" doesn't exactly apply. Finding it out may be hard, using it certainly is easy. It's not like you have to change anything about it to have a (more or less accurate) fast sqrt function.

      You CAN try to understand it (it's not that "magic" once you see why this applies), but you don't NEED to. It's a bit like DirectX coding. You CAN try to understand how DirectX does all the "magic" for you (like, say, calculate the normals, which this code is relevant to), but you don't NEED to anymore. DX is nice to you and just abstracts it all away. So maybe this (or a similar) hack is used to get fast sqrts. You just don't see them anymore.

      --
      We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
    12. Re:"Doom creator"? by Samantha+Wright · · Score: 2

      You raise a pretty good point, although you'd think they'd give him a slightly shinier honourific like "lead developer of the Quake engine".

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    13. Re:"Doom creator"? by _0xd0ad · · Score: 1

      Looks like a few things got eaten by the slashcode... that should be this:

      float Q_rsqrt( float number )
      {
              long i;
              float x2, y;
              const float threehalfs = 1.5F;
              x2 = number * 0.5F;
              y = number;
              i = * ( long * ) &y;
              i = 0x5f3759df - ( i >> 1 );
              y = * ( float * ) &i;
              y = y * ( threehalfs - ( x2 * y * y ) );
              return y;
      }

    14. Re:"Doom creator"? by morari · · Score: 1

      Fun fact: if it had been possible to make Doom like Tom Hall intended, storytelling in FPSes would be five years ahead from where it is today.

      Yeah, all of that storytelling worked well for Doom III...

      --
      "He who can destroy a thing, controls a thing." --Paul Atreides, Dune
    15. Re:"Doom creator"? by nabsltd · · Score: 1

      They look strange because the GP forgot to use HTML entities where needed, so some of the code disappeared.

      The original source (with comments) should explain things.

    16. Re:"Doom creator"? by Nick+Ives · · Score: 1

      Surely it should be "John Carmack is a creator of Doom."?

      --
      Nick
    17. Re:"Doom creator"? by paraduma · · Score: 1

      John Carmack used to post here :(

    18. Re:"Doom creator"? by Max+Romantschuk · · Score: 1

      +1 To you Sir. I was just about to comment the same thing.

      I have to say that ever if you're not a passionate gamer it's not that hand to Google it... ;)

      Educating oneself about things like these on the Internet is ultimately pretty easy compared to a lot of other settings.

      --
      .: Max Romantschuk :: http://max.romantschuk.fi/
    19. Re:"Doom creator"? by Max+Romantschuk · · Score: 1

      The real question is, why isn't Keen remade on WiiWare/PS store/Xbox live already?

      --
      .: Max Romantschuk :: http://max.romantschuk.fi/
    20. Re:"Doom creator"? by Rei · · Score: 1

      I actually dealt with a variety of fast square root and inverse square root functions and profiled them when creating a particle system for a MMORPG (Eternal Lands). That algorithm is indeed fast (at the cost of accuracy)... but it's still significantly slower than the hardware SSE inverse square root routine. Hence this function:


          static float invsqrt(const float f)
          {
      #ifdef __SSE__ // Hardware fast invsqrt.
              typedef union match { __m128 m128; struct { float x, y, z, w; }; } match;
              match f2;
              f2.m128 = _mm_rsqrt_ss(_mm_set_ss(f));
              return f2.x;
      #else // Quake fast invsqrt.
              typedef union match { int i; float f; } match;
              match tmp;
              float half = 0.5f * f;
              tmp.f = f;
              tmp.i = 0x5f3759df - (tmp.i >> 1);
              f = tmp.f;
              f = f * (1.5f - half * f * f);
              return f;
      #endif
          };

      I also found that caching values and interpolating between them for the pow() function can get a relevant speed increase (at notable cost to accuracy -- but for most 3d uses of pow, you don't care), but my efforts creating a sin cache didn't pay off, as it didn't improve performance any relevant amount.

      But back to the original issue: don't forget about __SSE__ (and later) instructions! There's some great ones, but a lot of the benefit only comes if you implement your class's members as __SSE__ (or other) data structures, bundling together multiple variables which are frequently used together. The performance increase on base classes (say, 3d vectors or RGBA colors) can be huge. The key is that you still have to have a naive implementation for hardware that doesn't whatever extensions you use. A good library will have half a dozen or more implementations of such a class for different hardware.

      One of the many reasons why it's best to not reinvent the wheel and just use existing library functions where available; widely used libraries already take care of all of this stuff for you.

      --
      He's just being nice so my real father won't freeze him in carbonite and sell him for spice.
    21. Re:"Doom creator"? by slew · · Score: 1

      Parent poster forgot about html ampersand escapes in thep copied it wrong...
      see missing semicolons at the end of a line? they should have & *y ; and & *i ;

    22. Re:"Doom creator"? by nomadic · · Score: 1

      Because it's not especially good?

    23. Re:"Doom creator"? by Samantha+Wright · · Score: 1

      They're actually pretty close (here's the original Doom Bible for comparison.) The point, though, is that FPSes without plot would have been way less established, and games of the mid-nineties would have gotten something of a better start in being story-bearing things. It would have been cheap science fiction, sure, but at least the story would have been action-movie-grade and not cereal-box-grade.

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    24. Re:"Doom creator"? by PaladinAlpha · · Score: 2

      Yeah, problems always look easy after someone else has solved them for you, I guess.

      Your advocacy of blissfully ignoring the internals fits in well with all the "yay Visual Studio!" and "yay .NET!" that I'm hearing in these comments; a legion of dull-eyed "developers" that are perfectly happy to eat only the food provided them, trusting their master to give them the right balance of nutrients. I'm sorry if that sounds harsh, but this whole thing is disgusting.

      How much better is Direct3D on Linux? How's that fork of Direct3D coming? When you leave state of the art to the corporations, and the corporations have huge marketshare, innovation dies. Is that what we want? Are we, after all the struggles to get here, content to lay under this particular tree until we rot?

    25. Re:"Doom creator"? by Anonymous Coward · · Score: 0

      That was John Romero. John Carmack had nothing to do with Daikatana, apart from developing the game engine and licensing it to Ion Storm.

    26. Re:"Doom creator"? by Anonymous Coward · · Score: 0

      Especially since the jerk who brought us Daikatana was John Romero, not John Carmack. He did cofound id though.

    27. Re:"Doom creator"? by Hatta · · Score: 2

      Doom is still the finest FPS ever created. A good arcade action game will stand the test of time. Like Tetris, Galaga, PacMan, etc Doom is a classic that will be played for decades, if not centuries. If you play these games for the story, you're missing the point entirely.

      The emphasis modern shooters place on story and graphics as opposed to level design and gameplay is a major reason they suck.

      --
      Give me Classic Slashdot or give me death!
    28. Re:"Doom creator"? by Anonymous Coward · · Score: 0

      That's John Romero

    29. Re:"Doom creator"? by Anonymous Coward · · Score: 0

      I'm on /. for the first time in probably 6 months. I used to be a regular here around 2000-2003. Boy has it changed and not for the better. I guess there was a reason I hadn't been here for a while.... dumbed down and becoming less relevant.

    30. Re:"Doom creator"? by Joe+Jordan · · Score: 1

      Indeed. Things sure have changed around here.

      For reference, check out this /. Q&A with Carmack in 1999: http://slashdot.org/story/99/10/15/1012230/John-Carmack-Answers

      One miscellaneous observation; quickly scroll through the thread. Posters put a lot more into their comments back then.

    31. Re:"Doom creator"? by vitaflo · · Score: 1

      The topic is about comparing one API to another.

      In that context, even if you were to go into a semantic rant about it, John Carmack is the creator of Doom.

      I read it as was why the need to describe who he is at all. On a site for nerds, just using the name John Carmack should be good enough for everyone to know who it is. It's akin to referring to Woz as "Apple creator".

    32. Re:"Doom creator"? by tibit · · Score: 1

      Thanks to wine developers, Direct3D on Linux is actually pretty darn good, all things considered.

      --
      A successful API design takes a mixture of software design and pedagogy.
    33. Re:"Doom creator"? by Samantha+Wright · · Score: 1

      As an experienced mapper I can say with some confidence that Doom failed to even meet its "arcade action game" goals: they took out the score system and treasure pickups shortly before release. You may want to turn your attention towards Wolfenstein 3-D or Duke Nukem 3D for that title; personally I always thought Blake Stone was a pretty memorable entry.

      But we all know the true god is Rise of the Triad—which, incidentally, was designed by Tom Hall.

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    34. Re:"Doom creator"? by Hatta · · Score: 1

      Played all of those. Doom is still the one I'll return to over and over again.

      --
      Give me Classic Slashdot or give me death!
    35. Re:"Doom creator"? by Max+Romantschuk · · Score: 1

      Because it's not especially good?

      I have to disagree. Keen 4-6 are quite good, but definitely challenging to play. Perhaps too much for some?

      --
      .: Max Romantschuk :: http://max.romantschuk.fi/
    36. Re:"Doom creator"? by Bacon+Bits · · Score: 1

      That's because Doom is the last game J.C. created. Look at the list of games since then:

      Dec 1993: Doom
      Oct 1994: Doom II. More Doom levels.
      Dec 1994: Heretic. Doom with magic instead of guns.
      Oct 1995: Hexen. More Doom with magic. And some RPG elements.
      1996: Final Doom. More Doom levels.
      May 1996: Strife. Doom with RPG elements. Literally.
      Jun 1996: Quake. Yeah, this is just Doom with aliens
      Mar 1997: Doom 64.
      Dec 1997: Quake II. More Doom with aliens
      Dec 1999: Quake III Arena. Doom: Deathmatch Edition.
      Dec 2000: Quake III Team Arena. Doom: Team Deathmatch Edition.
      Nov 2001: RtCW. Doom with Nazis.
      Aug 2004: Doom 3. Doom in the dark.
      Apr 2005: Doom 3: Resurrection of Evil. More Doom in the dark.
      Sep 2005: Doom RPG. It's, uh, Doom with RPG elements.
      Oct 2005: Quake 4. More Doom with aliens.
      May 2006: Orcs & Elves. Wow! It's new! Oh, wait, it's Doom RPG rebranded.
      Sep 2007: Enemy Territory: Quake Wars. Doom: Team Deathmatch Edition 2!
      Aug 2009: Wolfenstein. Doom with Nazis.
      Sep 2011: Rage. Post-apocalyptic Doom.
      Forthcoming: Doom 4. Yeah.....

      Don't get me wrong, the guy is a game engine genius. Still, he's very much a one-trick pony. He's the John Carpenter of video games rather than the Stanley Kubrick.

      --
      The road to tyranny has always been paved with claims of necessity.
    37. Re:"Doom creator"? by Clsid · · Score: 1

      I would like to thank you for sharing this piece of code. I'm always fascinated by computer code optimizations, even if D3D has it incorporated and OpenGL doesn't have it, it's good to know the internals.

    38. Re:"Doom creator"? by billcopc · · Score: 1

      Factually speaking, it's not exactly like we can say Bobby Prince wrote anything. Just ask: Pantera, Slayer, Metallica, Black Sabbath, hell even that weak jesus-metal band Believer. Pretty much the entire Doom soundtrack consisted of FM covers of commercial metal tracks.

      --
      -Billco, Fnarg.com
    39. Re:"Doom creator"? by billcopc · · Score: 1

      Dude, I see you have a 5-digit ID. I know it's hard to accept for us old-timers, if you still believe /. is "a site for nerds", you need to lay off the smack. The last real nerds bailed, right around the time they made Jon Katz an editor. This place is dead, it just doesn't know it yet.

      --
      -Billco, Fnarg.com
    40. Re:"Doom creator"? by MikeBabcock · · Score: 1

      Heretic and Hexen were much better games anyway.

      --
      - Michael T. Babcock (Yes, I blog)
    41. Re:"Doom creator"? by MikeBabcock · · Score: 1

      Mmm, zombies.

      My apologies for the one-line reply.

      Oh crap, now it isn't.

      --
      - Michael T. Babcock (Yes, I blog)
    42. Re:"Doom creator"? by Samantha+Wright · · Score: 1

      Oh, he wrote plenty of stuff—like that timeless powerhouse of 20th Century composition, Eat Your Veggies.

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    43. Re:"Doom creator"? by Samantha+Wright · · Score: 1

      Bringing Raven Software into this is like bringing a broadsword to a knife fight.

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    44. Re:"Doom creator"? by Opportunist · · Score: 1

      Sadly, while absolutely correct from a technical point of view, you are wrong from an economic point of view. I'd prefer a "real" programmer, one who knows WHY he does something, over a rote coder, someone who often just copy/pastes snippets that (again, often only SEEM TO) do what he wants to do and relies on libraries for the rest. Unfortunately, the economy does not. Economy cares about mandays per result, and in this field the rote coder often wins. His solution will certainly be inferior, it will be slower, more error prone, probably hard or impossible to scale reasonably and maybe even hard or impossible to maintain. But it will be one thing: It will be delivered sooner. It seems everything managers hear about best practice in their beloved certification courses goes out the window the moment they notice that they could cut a corner and increase their bonus check.

      Hence RAD will always win out over sensible planning, implementation and testing. And hence we got rote coders with zero to little insight concerning the whys of programming, the underlying mathematic principles, hell, even mentioning Big O-Notation to some database coders earns you blank stares. And when you mention Landau they'll probably think you want to talk about movies, at least if they're into cheesy scifi. And they are right, from an economic point of view. It's very often cheaper to buy faster machines than to improve the queries or the algorithm.

      Know your audience. And, as a programmer, your audience is usually NOT the customer. It's management.

      --
      We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
    45. Re:"Doom creator"? by morari · · Score: 1

      I don't know. I thought that Quake did a great job of telling it's story through sheer atmosphere if nothing else. Quake 2 did even more to build the meta-plot through it's mission updates. It didn't take much to move past Doom.

      --
      "He who can destroy a thing, controls a thing." --Paul Atreides, Dune
    46. Re:"Doom creator"? by Samantha+Wright · · Score: 1

      That's my point exactly. Doom as originally envisioned by Tom Hall would stand up to Half-Life 1 in terms of plot substance.

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    47. Re:"Doom creator"? by morari · · Score: 1

      I don't doubt that. I do however wonder if the technology of the time could have truly presented what was envisioned. I don't think it could have, not adequately at least.

      --
      "He who can destroy a thing, controls a thing." --Paul Atreides, Dune
    48. Re:"Doom creator"? by Samantha+Wright · · Score: 1

      What we know puts pretty difficult-to-envision expectations on the game, true—Hall was hoping for his full-blown first episode to fit on a floppy, when the finished and much less ambitious demo was a solid 5 megs. But most of the plot was going to be through text messages (like in Catacombs 3-D, the first 3D game made by Carmack/Carmack/Hall/Romero) and cutscenes; and an appropriately advanced cutscene engine was included in RotT a year later (but barely utilized). It would have been possible technically, but not with the resources id Software had at the time, and they would probably have been beaten in the market by early BUILD games like TekWar and Witchhaven (in fact, it's possible that Duke Nukem 3D would have been on the market, too, albeit as a very different game—a direct sequel to Duke Nukem 2.)

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    49. Re:"Doom creator"? by Rexdude · · Score: 1

      That was John Romero, not Carmack.

      --
      "..One hosts to look them up, one DNS to find them, and in the darkness BIND them."
    50. Re:"Doom creator"? by scot4875 · · Score: 1

      That said, Id hasn't released an interesting game since Romero and McGee left.

      --Jeremy

      --
      Jesus was a liberal
    51. Re:"Doom creator"? by nomadic · · Score: 1

      Seemed like a standard platformer to me. Challenging doesn't have much impact on my opinion of a game; I've played easy games that I thought were amazing (Loom) and ridiculously hard games that in effect beat me because I couldn't get through them that were amazing (the Dreamcast Ecco the Dolphin game).

  4. Oh my.... by xQuarkDS9x · · Score: 1

    I never thought I'd see the day John Carmack would admit DirectX is better then OpenGL. I remember since the early days of doom and quake how they trumpeted OPENGL as the end all and be all of the flagship games. I guess miracles do happen!

    --
    You must master your joystick like a fisherman masters bait! - Gimpy
    1. Re:Oh my.... by Osgeld · · Score: 1

      cept he has been quoted on this before, headline should read

      "carmack still thinks the same" cripes its even in a wikipedia entry

      Some former critics of Direct3D acknowledge that now Direct3D is as good if not better than OpenGL in terms of capabilities and ease of use. In January 2007, John Carmack said that "DX9 is really quite a good API level. Even with the D3D side of things, where I know I have a long history of people thinking I'm antagonistic against it. Microsoft has done a very, very good job of sensibly evolving it at each step—they're not worried about breaking backwards compatibility—and it's a pretty clean API. I especially like the work I'm doing on the 360, and it's probably the best graphics API as far as a sensibly designed thing that I've worked with."[4]

      http://en.wikipedia.org/wiki/Comparison_of_OpenGL_and_Direct3D

    2. Re:Oh my.... by Anonymous Coward · · Score: 0

      We can thank Khronos for backing out on the OpenGL 3.0 redesign...

    3. Re:Oh my.... by Opportunist · · Score: 2

      It actually elevated him in my esteem. Nothing's worse than someone clinging to some technology akin to a sacred cow. If he deems it better and uses it because he considers it better, more power to him.

      A good programmer (and a good project lead) will use the tools best suited to the project. Not cling to his pet technology. And everyone who ever had to serve under someone who had one such pet tool/technology/language will know the value of project leaders who can look past their favorite toy.

      --
      We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
    4. Re:Oh my.... by morari · · Score: 1

      Yep. Now he can go back to making console-friendly games like RAGE, all under the guidance of his new corporate overlords at Bethesda. :(

      --
      "He who can destroy a thing, controls a thing." --Paul Atreides, Dune
    5. Re:Oh my.... by rfdparker2002 · · Score: 1

      yes sure a miracle - the chance of more great games/engines being tied to a single vendor's products. Yes OpenGL has issues but if Microsoft really had an interest in having a good cross-platform API, they could surely help move OpenGL forward in appropriate ways rather than putting that effort into developing something else entirely.

    6. Re:Oh my.... by Bengie · · Score: 1

      I find it funny that many Wine developers said D3D11 is easier and more structured than OpenGL. Even OpenSource developers like DX over OpenGL.

    7. Re:Oh my.... by Bengie · · Score: 1

      I can't elevate my esteem(in computing) for Carmack without overflowing back to 0. He is a genius in his field.

    8. Re:Oh my.... by MikeBabcock · · Score: 1

      I can't elevate my esteem(in computing) for Carmack without overflowing back to 0. He is a genius in his field.

      QFT.

      He's also a genius in a few other fields I suspect.

      --
      - Michael T. Babcock (Yes, I blog)
  5. WTH? by Ramirozz · · Score: 0

    "Microsoft had the courage to continue making significant incompatible changes to improve the API" apply this to politics and it gets a little scary... oh... wait.

    --
    http://www.quasarcr.com/
    1. Re:WTH? by YodasEvilTwin · · Score: 1

      What the hell does that have to do with anything?

  6. Doom Creator? by Anonymous Coward · · Score: 0

    Pretty sure you don't need to refer to Carmack as the 'Doom Creator' I mean, this is /. for crying out loud ;)

  7. Hometown Hero by Anonymous Coward · · Score: 0

    We, the citizens of Mesquite, TX, thank you John Carmack for revealing the obvious. Oh, and Sony makes shitty stereos.

  8. An extension of engine recycling by Anonymous Coward · · Score: 0

    I think John has a good point - the Quake engine has drove a lot of games, along with the Unreal engine. Valve's flagship, the Source engine, is a fork of the quake engine if memory serves. Back in the day, Quake 2 and Unreal / Unreal Tournament had OpenGL support, but that's around the point that I started seeing more games using DirectX, and now we have the big commercial engines using it exclusively that other developers license for their own games.

    Would DirectX have this market if more of the big engines used OpenGL?

    1. Re:An extension of engine recycling by VGPowerlord · · Score: 2

      Valve's flagship, the Source engine, is a fork of the quake engine if memory serves.

      GoldSrc (the Half-Life 1 engine) was a fork of the Quake 1 engine. Source is a continuously upgraded version of the GoldSrc engine. The most recent version of the engine in a released game is the Left 4 Dead 2 engine, which added a few new features that weren't present in earlier version, such as the AI Director and the fog simulation.

      Having said that, at some point OpenGL was dropped from Source, and finally re-added when the OSX version shipped.

      I've heard that EA also added OpenGL support when they ported Orange Box to the PS3, but I don't think Valve reincorporated that into their own games, but instead wrote the OpenGL support from the ground up.

      --
      GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
    2. Re:An extension of engine recycling by Anonymous Coward · · Score: 0

      Wouldn't it be more correct to say that TF2 is the newest version of the Source engine? When you get a TF2 update, DODS, HL2DM, CSS, and sometimes the episodes get an update. L4D(2) are dead end ( heh, dead + zombies ) branches of the Source engine rather than main versions.

    3. Re:An extension of engine recycling by Beelzebud · · Score: 1

      When L4D2 shipped it was, in fact, using the latest version of the Source engine. At some point they updated TF2 and the HL2 episodes to that version.

  9. What shocks me the most.. by Viol8 · · Score: 2

    ... is how old Carmack looks now! I still remember him from the Doom days and I haven't seen a picture since. Came as a bit of a shock.

    1. Re:What shocks me the most.. by 93+Escort+Wagon · · Score: 2

      ... is how old Carmack looks now! I still remember him from the Doom days and I haven't seen a picture since. Came as a bit of a shock.

      Good grief - reading your comment, a person might think he now looked like Gandalf!

      He's just middle-aged, it happens to everybody.

      --
      #DeleteChrome
    2. Re:What shocks me the most.. by Sycraft-fu · · Score: 3, Insightful

      And it is far preferable to the alternative.

    3. Re:What shocks me the most.. by geekoid · · Score: 1

      Being young forever?

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    4. Re:What shocks me the most.. by monoqlith · · Score: 1

      Better Gandalf than Gadaffi.

    5. Re:What shocks me the most.. by HornWumpus · · Score: 1

      Young and dead.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    6. Re:What shocks me the most.. by tyrione · · Score: 1

      He looks ten years my senior and he's two years younger. He was never known for his exterior charisma.

      http://en.wikipedia.org/wiki/File:John_Carmack_GDC_2010.jpg

    7. Re:What shocks me the most.. by Anonymous Coward · · Score: 0

      I'm 25, but I looked at his picture and thought, "This is the guy who *created* a game I loved back when I was just tall enough to mash keys our church's 386! He should be older..."

      He reminds me of Dave from NewsRadio... The character who gets made fun of constantly for looking like a 14 year old.

    8. Re:What shocks me the most.. by Anonymous Coward · · Score: 0

      dragon age?

  10. Not only that by Sycraft-fu · · Score: 5, Insightful

    They work with the GPU manufacturers. Basically when new GPUs are in development, so is the new DirectX. So MS has a chat with nVidia and AMD. They tell the GPU makers the kind of things they want, the GPU makers tell them the kind of things there hardware is going to have, and they are able to come to a standard that everyone supports. That is why when new GPUs come out they support all the features of the new DX. It isn't some amazing coincidence. Also it is proper support, a single standard that works well with the abilities the cards have. You write your DX driver, and everything works.

    OpenGL functions in much more of a lagging capacity. New video cards come out, and then it gets support for whatever it is they bring to the table sometime later. Khronos doesn't seem to go out and engage the vendors during development and try to have OpenGL ready to meet the next gen cards. Also their strategy often seems to be "just use extensions for it," which means that you can have differences between vendors for how things work.

    1. Re:Not only that by Anonymous Coward · · Score: 1

      The problem with the OpenGL ARB is that it's design-by-clusterfuck. Instead of Khronos Group laying out specs along the lines of "here's the draft, we want your opinions about whether you can make it or not", it's "Vendor X suggests method X1, vendor Y suggests method Y1, 2 weeks of shitslinging goes on, suggestions modified, rinse and repeat".

      And even up until 2006, ATI were actively working against OpenGL, they had all their focus on DirectX(In fact, all the way up until the Radeon 9800 Pro, none of their consumer cards would manage to properly run SpecViewPerf, and even when the 9800 Pro managed, it was BAD. Even the FireGL's would crash often, or render garbage.

      That's not to say that ATI's DirectX was that much better... it was speedy.... but there was no end to the amount of "fixed graphics issue nr gazzillion, which would affect ATI users" in patches for games.

    2. Re:Not only that by Rei · · Score: 1

      DirectX, OpenGL.... why should the end user care? Unless you're developing a rendering engine, you should probably just be using a rendering engine not caring about the underlying mechanics. I mean, how many times should we have to implement stencil shadows from scratch, mesh loading from scratch, paged geometry from scratch, and on and on? Odds are, if more than just a handful of people need a piece of functionality, a good rendering engine probably already has it implemented -- and a better, faster, more capable, more bugfree version than you'd write. And if not? Write it on your own and then contribute your code.

      It's like using STL versus not. Sure, you *could* go and implement STL container functionality on your own, but why would you want to?

      --
      He's just being nice so my real father won't freeze him in carbonite and sell him for spice.
    3. Re:Not only that by Anonymous Coward · · Score: 0

      Most of the new 3D tech of the past ten years or so was available on OpenGL first. As extensions, yes, and that does mean that you have to account for the possibility that the extension is unavailable, but then you cannot use the newest Direct3D on old hardware, so yeah. In any case, bringing up one of the few exceptions doesn't prove much of a point.
      Neither does the multi-threading thing.
      And Direct3D's multi-threading support isn't anything fancy either. Nothing you cannot do yourself in OpenGL, and you still and up doing most of the synchronisation yourself.
      In short, "Commander Keen creator says Direct3D is now better than OpenGL" is an argument from authority. Arguments from authority are fallacious since both the reasoning and the premises of the argument may be wrong, and that the argument was made by a famous person doesn't change that.

    4. Re:Not only that by turgid · · Score: 2

      DirectX, OpenGL.... why should the end user care?

      Because I don't want to have to buy a Microsoft-installed computer (expensive, restrictive, unreliable. insecure, user-unfriendly) just to do some 3D stuff.

      On what other operating systems does Direct3D run? OpenGL, for all its faults, is an Open Standard and there are several compatible, competing implementations including Open Source ones (Mesa).

      OK, so I'm a loony and the doctor gives me pills...

    5. Re:Not only that by Rei · · Score: 1

      You know, it helps if you don't stop reading a post after the first sentence.

      --
      He's just being nice so my real father won't freeze him in carbonite and sell him for spice.
    6. Re:Not only that by bhcompy · · Score: 1

      And this is why 3dfx went the Glide route. OpenGL was a clusterfark, so they made their own GL like/based API that kicked ass and was easy to develop for

    7. Re:Not only that by dwightk · · Score: 1

      that went well for them :D

      --
      Like anyone can even know that
    8. Re:Not only that by MBGMorden · · Score: 1

      In all actuality it did. Back in those days pretty much all games were built for Glide. If you didn't have a Voodoo card you didn't REALLY have a 3d accelerator. Eventually when other hardware started catching up and a more general purpose API was needed, 3dfx supported those without issue.

      EVENTUALLY poor hardware and a poor decision to go to first-party manufactured cards killed 3dfx, but Glide certainly wasn't one of their mess ups.

      --
      "People who think they know everything are very annoying to those of us who do."-Mark Twain
    9. Re:Not only that by MBGMorden · · Score: 1

      DirectX, OpenGL.... why should the end user care?

      Because in the context of this discussion, game developers ARE the end user. You're doing the equivalent of going to a carpenter's forum where they're discussing which brand of hammer is better and then screaming that the home owner isn't going to care. That's all well and good, but the home owner isn't the user of the products being discussed.

      --
      "People who think they know everything are very annoying to those of us who do."-Mark Twain
    10. Re:Not only that by Josh04 · · Score: 1

      In short, "Commander Keen creator says Direct3D is now better than OpenGL" is an argument from authority. Arguments from authority are fallacious since both the reasoning and the premises of the argument may be wrong, and that the argument was made by a famous person doesn't change that.

      I think you'll find it's not actually an argument at all. It's the title of an article linking to the argument, which you haven't read.

    11. Re:Not only that by Anonymous Coward · · Score: 5, Informative

      [quote] ...Khronos doesn't seem to go out and engage the vendors during development and try to have OpenGL ready to meet the next gen cards. ....
      [/quote]

      Complete utter bullshit. The chair of the ARB OpenGL working group is the lead for GL driver development for NVIDIA. Much of the jazz for the OpenGL specifications is driven by the hardware makers. Many GL extensions that were first exposed by NVIDIA made it into GL3, and that tradition continues today.

      The major stinks people have had with GL:
        1) GL3 was late and a disappointment (the disappointment is mostly related to not getting direct state access of GL objects)
        2) Direct State Access (i.e. can edit objects without binding them first) is available as an extension. As of now AMD and NVIDIA support that extension (it has other parts too, for reference that extension's name is GL_EXT_direct_state_access. I'd make a substantial bet that GL4.2 will have it in core.
        3) Few tools in comparison to Direct3D (though glDebugger is for free now)
        4) Inconsistent behavior and implementations of GL across different hardware vendors. Intel's GL sucks monkey balls, NVIDIA's lets you get away with murder and ATI makes you follow the spec with all i's dotted and t's crossed. Additionally, what is fast for one IHV is not fast for another (buffer object usage hints I am looking right at you).

      There is also some truth to that GL now "follows" Direct3D, i.e GL4 came out after D3D11, etc.
      Though, there are extensions in GL that completely blow the living crap out of anything you can do with Direct3D, take a look at GL_NV_shader_buffer_load. Full blown (read only) pointers in shaders (pointers point to data of buffer objects). Nothing in Direct3D compares to it.. and it is for GeForce 8/9/1xx/2xx/3xx cards! For the next generation (GeForce4xx) there is even write access (GL_NV_shader_buffer_store).

    12. Re:Not only that by tibit · · Score: 1

      This should be modded +5 Informative. It seems to be the only technical, to-the-point post in the whole thread. Kudos!

      --
      A successful API design takes a mixture of software design and pedagogy.
    13. Re:Not only that by Anonymous Coward · · Score: 0

      This. Plus the fact that Intel chipsets use Mesa drivers, which are stuck in the OpenGL 2.1 stone ages.

      Want to write an OpenGL 3.x app for mass consumption? Sorry -- you can't do it until Mesa releases an OpenGL 3.x driver, because most "casual" gamers have Intel chipsets that are powerful enough and perfectly capable of doing OpenGL 3.x functionality, but they can't do it until Mesa releases a 3.x driver.

      • 3.0 adds frame buffer objects, conditional rendering, vertex array objects, floating point textures and floating point depth buffers.
      • 3.1 adds instancing and uniform buffer objects, and makes it possible to access 16 textures from the vertex shader.
      • 3.2 includes (non-ARB) geometry shader, plus sync/fence objects.

      If you want modern eye candy, you need instanced rendering and a geometry shader for rendering tens of thousands of plants, plus floating point depth buffer for deferred shadow rendering. Try doing that with OpenGL 2.1.

    14. Re:Not only that by Anonymous Coward · · Score: 0

      well, i dont think you are quite right, AFAIK Khronos consists of the graphics vendors(+others).
      Of course, Microsoft approaches the vendors, but they also consort with the OpenGL developers, thats why when new graphics cards come out, they DO support all OpenGL features of the latest standard(at least nvidia does, idk about amd).

      In my experience, the problem with OpenGL is in fact that they care(d) about backwards compatibility, which makes the whole system inflexible and static; this ,IMO, is totally unsuitable for the highly innovative computer graphics area where new features and techniques to increase performance are critical.

      Thankfully Khronos has started to release new OpenGL specs on a regular basis since OpenGL 3.0 and is making a lot of progress now. Of course, there is still much to catch up, but I am optimistic about the future of OpenGL because it seems that Khronos does actually thhttp://games.slashdot.org/story/11/03/11/1832205/Doom-Creator-Says-Direct3D-Is-Now-Better-Than-OpenGL#rive to bring at least a bit of innovation to OpenGL now.

    15. Re:Not only that by turgid · · Score: 1

      You know, it helps if you don't stop reading a post after the first sentence.

      What are you on about? I'm an "end user" of the graphics library.

      You, know, you're right. after all. I don't play games nowadays, so why the heck would I want 3D? I'm 36 years old. It's been 10 years since I bought a game. I suppose it doesn't matter if no one but Microsoft customers can run hardware-accelerated 3D code on their computers.

    16. Re:Not only that by Kjella · · Score: 1

      DirectX, OpenGL.... why should the end user care?

      The end user picks the games that are good.
      The games pick the engines that are good.
      The engines pick the interfaces that are good.

      So if the engines find DirectX is best, that's what they bring fastest to market with best performance and least bugs then it cascades down to the user having a lot of DirectX games to pick from. Which of course means Windows, unless you want to mess with WINE. Also games do care about code quality, if the average OpenGL implementation is buggier and less optimized than the DirectX implementation that would reflect poorly on the game. So it doesn't matter to the end user, but it matters to the people who'll decide what the end user has to choose from.

      --
      Live today, because you never know what tomorrow brings
    17. Re:Not only that by Prune · · Score: 1

      This is no longer true. OpenGL support is now apace with any new hardware features, making them available as OpenGL extensions, basically _EXT. The extensions are available to the developer immediately upon installation of the appropriate driver--and at least NVIDIA driver releases support all features of the hardware immediately).

      --
      "Politicians and diapers must be changed often, and for the same reason."
    18. Re:Not only that by Prune · · Score: 1

      Damn slashdot removed my < and > ... Why do we have to use HTML for input? It should be an option, not mandated. This is why we can't have nice things -.-
      Replace _EXT in my post above with <function/define name>_EXT

      --
      "Politicians and diapers must be changed often, and for the same reason."
    19. Re:Not only that by returnofjdub · · Score: 3, Interesting

      This post doesn't make any sense. The people who define the OpenGL spec include delegates from ATI, Creative Labs, Intel, and Nvidia. Khronos doesn't go out and engage the vendors? They're a consortium of the vendors you claim they don't engage.

      The reason for the core OpenGL spec lagging with consumer level graphics stuff is largely due to its incredible breadth of applications and target platforms. With OpenGL 3.0 there was a lot of contention between the people who wanted to turn it into a streamlined real time gaming API, and the people who used it for other industries where a lot of features not supported on gaming hardware were still useful for non-gaming applications. It started with a very ambitious revamping proposal, followed by months of (rather aggrivating) total silence, then culminated with the deprecation model that's currently in place.

      OpenGL's problems don't have anything to do with Direct3D being more in the loop than they are about new hardware developments. It's inherently more challenging to keep pace and be flexible because they're maintaining a broad spec used by a lot of different companies in a lot of different specialized fields. As far as feature deployment is concerned, on more than one occasion Nvidia has had drivers out on the same day Khronos releases a new spec. Current desktop OpenGL is quite a nice, modern API that's suitable for cutting edge game development. The new deprecation model allows driver authors to create profiles optimized for specific classes of applications (of which there are many where OpenGL is useful). On Windows, libraries like GLEW make Microsoft's decision to not move the ABI forward a non-issue. OpenGL ES is very much a modern API that's useful across a wide variety of in-demand mobile platforms. The way modern GL handles things like VBOs and render to texture are at least as good as Direct3D.

      I'm not advocating the use of OpenGL over D3D or vice versa. Right now I primarily do Android and web development stuff, so I'm kind of saturated in OpenGL-centric environments. I just felt the need to respond to that weird claim that Khronos is disengaged from hardware vendors. Khronos largely ARE hardware vendors.

      Sincerely,
      MS Fanboy with an Xbox 360, XNA Creators Club subscription, and a deep love for Visual Studio and C# who uses Bing search (and doesn't think IE9 is absolutely terrible)

    20. Re:Not only that by Daniel+Phillips · · Score: 1

      The problem with the OpenGL ARB is that it's design-by-clusterfuck. Instead of Khronos Group laying out specs along the lines of "here's the draft, we want your opinions about whether you can make it or not", it's "Vendor X suggests method X1, vendor Y suggests method Y1, 2 weeks of shitslinging goes on, suggestions modified, rinse and repeat".

      Well, but the ARB specifications typically end up in the excellent zone, so something must be working. I actually read them, do you?

      --
      Have you got your LWN subscription yet?
    21. Re:Not only that by Daniel+Phillips · · Score: 1

      how many times should we have to implement stencil shadows from scratch

      Zero. Stencil shadows suck beyond belief, just don't do it.

      --
      Have you got your LWN subscription yet?
    22. Re:Not only that by Daniel+Phillips · · Score: 1

      "Commander Keen creator says Direct3D is now better than OpenGL"

      Heh, nice summary. John Carmack remains the preeminent optimizer of the game world, however the game world has pretty much passed optimizers by in favor of people who can deliver compelling content quickly. Actually, what he really said is, DirectX is marginally better than OpenGL (which is not at all clearly true) but the difference is marginal so it's not worth rewriting all his editor code. Nice way to blow tons of karma. Obviously, he isn't going to be implementing iPhone Rage on DirectX any time soon.

      --
      Have you got your LWN subscription yet?
    23. Re:Not only that by Daniel+Phillips · · Score: 1

      1) GL3 was late and a disappointment (the disappointment is mostly related to not getting direct state access of GL objects)

      I was elated and relieved. The last thing we needed is to lose backward compatibility. The way it was finally done, with compatibility profiles, is perfect. A half cocked solution should have truly been the death of OpenGL. Major kudos to those level heads who prevailed.

      --
      Have you got your LWN subscription yet?
    24. Re:Not only that by Daniel+Phillips · · Score: 1

      Intel chipsets use Mesa drivers, which are stuck in the OpenGL 2.1 stone ages.

      Mesa @2.1 isn't too bad with all the extensions. But yes, it does lag. I think it was only recently I could use framebuffer objects and I absolutely refuse to write code to fall back to pbuffers.

      --
      Have you got your LWN subscription yet?
    25. Re:Not only that by Daniel+Phillips · · Score: 1

      In my experience, the problem with OpenGL is in fact that they care(d) about backwards compatibility, which makes the whole system inflexible and static; this ,IMO, is totally unsuitable for the highly innovative computer graphics area where new features and techniques to increase performance are critical.

      It is easy to understand the importance of backward compatibility. Do you really want your 3D apps to stop working because the libaries got updated?

      The compatibility profile solution that arrived in 3.1 is fine, well worth waiting for.

      --
      Have you got your LWN subscription yet?
    26. Re:Not only that by MikeBabcock · · Score: 1

      Did them just fine now that they're part of NVidia.

      --
      - Michael T. Babcock (Yes, I blog)
    27. Re:Not only that by MikeBabcock · · Score: 1

      I always enjoy the view that OpenGL is dead or dying when its the only usable 3D API for industrial and mobile applications. PC Games aren't exactly the entire world of 3D.

      --
      - Michael T. Babcock (Yes, I blog)
    28. Re:Not only that by TheTurtlesMoves · · Score: 1

      The games pick the engines that are good.

      Not quite. They pick engines that are good and *run* on the target platform/s, and that target platform is *popular*. Good platforms are useless if no one owns it. Once a platform is chosen, for the most part so is the interface.

      Also these days piracy concerns and perceptions does drive platform choice.

      While i do a bit of 3d, and i don't just use a Xbox or windows, the choice of interface is pretty small. Basically OpenGL is my only choice. But i am happy with it, last time i used DX (9.3a IIRC) it was very similar. In both cases most of your work is driver bug workarounds.

      --
      The Grey Goo disaster happened 3 billion years ago. This rock is covered in self replicating machines!
    29. Re:Not only that by TheTurtlesMoves · · Score: 1

      I could not agree more. The game forum web sites went ape over GL3. But it seems the only folks blowing chunks were people that clearly didn't work with OpenGL or didn't work with it much. Most of my friends that are working on CAD tools were relived with the opengl3 spec, and a totally stoked with how fast opengl is moving now.

      --
      The Grey Goo disaster happened 3 billion years ago. This rock is covered in self replicating machines!
    30. Re:Not only that by TheTurtlesMoves · · Score: 1

      For some people it is. The irony is that all the graphics was probably don't in a 3d program that used opengl. I know that most CAD only use opengl and I know at least one major 3d graphics program does too.

      --
      The Grey Goo disaster happened 3 billion years ago. This rock is covered in self replicating machines!
    31. Re:Not only that by __aailob1448 · · Score: 1

      So basically, openGL is to direct3d what the gimp is to photoshop.

    32. Re:Not only that by Lord_Jeremy · · Score: 1

      As far as I've seen, game developers working on their game engines re-implement everything themselves. The Source engine is written in C++, but makes virtually no usage of any STL. They wrote their own container implementations, etc. The why would you want to probably stems from the desire to have full control over the implementation of all of their "under the hood" stuff. I can certainly think of a few situations when I wish I could make some change to the STL map class or somesuch. Not to mention in large development project you probably want to avoid dependance on external libraries wherever possible.

    33. Re:Not only that by Rei · · Score: 1

      Do you know what a rendering engine is? Look it up before you respond to a post. Developers shouldn't be developing to either Direct3D *OR* OpenGL. They should be developing to a rendering engine and only care what platforms the rendering engine supports. The rendering engine should take care of the implementation (good ones run on both OpenGL and Direct3D).

      --
      He's just being nice so my real father won't freeze him in carbonite and sell him for spice.
    34. Re:Not only that by Rei · · Score: 1

      No, what I'm doing is the equivalent of 'screaming' that the carpenter should be using a nail gun.

      The "carpenter" (programmer) shouldn't be using OpenGL, and they shouldn't be using Direct3D. They should be using a rendering engine. CrystalSpace, TrueVision, OGRE, Irrlicht, CryEngine, Unreal Engine, whatever.

      The fact that I'm having to bother to explain this is really sad. It means lots of people are out there reinventing the wheel.

      --
      He's just being nice so my real father won't freeze him in carbonite and sell him for spice.
    35. Re:Not only that by Rei · · Score: 1

      The good game engines are cross-platform; this is either done by being OpenGL only or, much more common, supporting both OpenGL and DirectX. The only good one that I can think of that isn't is the CryEngine series. I'm sure there are others, mind you.

      --
      He's just being nice so my real father won't freeze him in carbonite and sell him for spice.
    36. Re:Not only that by Rei · · Score: 1

      It all depends on the situation. In a number of situations, stencil shadows look a lot better.

      Anyway, a good rendering engine lets you pick between shadow types.

      --
      He's just being nice so my real father won't freeze him in carbonite and sell him for spice.
    37. Re:Not only that by turgid · · Score: 1

      If someone as famous as Carmack comes out and says that Direct3D is "better than" OpenGL, and if the PHBs of all the big 3D hardware companies get it into their thick little skulls that Direct3D has won, you can be sure that support for OpenGL will start to be withdrawn for "good business reasons."

      Then everyone who doesn't have Windows will not be able to buy new 3D hardware and the OpenGL platform will die. We will be left with legacy hardware and bit-rotting software.

      How does writing to a rendering engine help?

    38. Re:Not only that by turgid · · Score: 1

      The fact that I'm having to bother to explain this is really sad. It means lots of people are out there reinventing the wheel.

      Reinventing the wheel can be fun and educational.

      It's not usually a good idea when working in a competitive business, though.

    39. Re:Not only that by MBGMorden · · Score: 1

      The reality though is that those engines still don't spring up out of the ground. They have to be written in the first place, and those programmers have to choose D3D or OpenGL. There is cause to care which operates better.

      Besides - existing engines aren't always the best choice. If no one ever reinvented the wheel they'd probably still be made of wood.

      --
      "People who think they know everything are very annoying to those of us who do."-Mark Twain
    40. Re:Not only that by badkarmadayaccount · · Score: 1

      Nothing serious - shift development to Direct3D state tracker in Gallium3D.

      --
      I know tobacco is bad for you, so I smoke weed with crack.
    41. Re:Not only that by turgid · · Score: 1

      Interesting.

  11. This being Slashdot... by 93+Escort+Wagon · · Score: 0

    I predict every comment that doesn't take issue with Carmack's statement will be modded down. So...

    OpenGL is obviously and demonstrably better than Direct3D. Not only is it open, it's Open! Direct3D only has ONE of those letters, and it's not even the important one - it's only the "e"!

    --
    #DeleteChrome
    1. Re:This being Slashdot... by 91degrees · · Score: 1

      Yes, but some of us prefer directness over openness:)

  12. AMD in passenger seat by Anonymous Coward · · Score: 0

    Well the interesting comment in that article is the one from AMD.

    'The actual innovation in graphics has definitely been driven by Microsoft in the last ten years or so,' explained AMD's GPU worldwide developer relations manager, Richard Huddy.

    One would imagine that a company that develop and make GPU accelerators would be the innovator in the field but apparently AMD is fine with being in Microsoft shadow.

    1. Re:AMD in passenger seat by KillAllNazis · · Score: 0

      They got paid to say that.

    2. Re:AMD in passenger seat by Urkki · · Score: 2

      Well the interesting comment in that article is the one from AMD.

      'The actual innovation in graphics has definitely been driven by Microsoft in the last ten years or so,' explained AMD's GPU worldwide developer relations manager, Richard Huddy.

      One would imagine that a company that develop and make GPU accelerators would be the innovator in the field but apparently AMD is fine with being in Microsoft shadow.

      Modern graphics hardware is nothing without a library and API to it. At least for gaming purposes (and excluding consoles, though they really aren't cutting edge by the time they're in the shops anyway), Microsoft controls what programmers can ask hardware to do for them, and therefore ultimately they control what hardware can be designed to do.

  13. The answers depend on the questions by SplashMyBandit · · Score: 5, Informative

    Carmack changed his mind some years ago. This report is quite late.

    However, the title of the magazine is "Custom PC". It is worth keeping in mind that if the PC and Xbox are the only platforms you are targeting then DirectX is a valid choice for development technology.

    Otherwise, you are better off developing in OpenGL, where you can target PCs, PS3, iPhone, iPad, Mac OS X, WebGL, industrial Unix (not all 3D apps are games, dontchaknow?). The only thing you can't do much with is the Xbox (technically possible, but deliberately closed by Microsoft).

    Also, the pace of change in OpenGL has picked up tremendously with the stewardship of the Khronos group. So OpenGL is starting to have parity in features again after lagging for some time (plus, you can get those features on Windows XP for those still on it).

    1. Re:The answers depend on the questions by Vegemeister · · Score: 1

      Protip: Macs and Linux boxen are PCs.

    2. Re:The answers depend on the questions by H0p313ss · · Score: 1

      Hint: This is not one of those places where being a pedant is actually helpful.

      --
      XML is a known as a key material required to create SMD: Software of Mass Destruction
    3. Re:The answers depend on the questions by Anonymous Coward · · Score: 0

      Not if they are servers. The 'P' means persona.

    4. Re:The answers depend on the questions by Narishma · · Score: 1

      You can remove the PS3 from that list as nobody uses OpenGL on it.

      --
      Mada mada dane.
    5. Re:The answers depend on the questions by nomadic · · Score: 1

      Ow. Do people use the term "boxen" still? I had hoped that unfortunate cutesy term had become extinct.

    6. Re:The answers depend on the questions by SplashMyBandit · · Score: 1

      Sure, they use toolkits instead. If they want do 'to the metal' optimizations though, they won't be using DirectX for it.

    7. Re:The answers depend on the questions by hedwards · · Score: 1

      That aspect is what's shocking. Carmack still ports things to random architectures to show that he's still the man. And a lot of those presumably don't have DirectX. Or at least I assume that the iPhone doesn't.

    8. Re:The answers depend on the questions by anss123 · · Score: 1

      I think his point was that they won't be using OpenGL either. Both Sony and Nintendo offer proprietary APIs for their consoles, you can use OpenGL - but from what I've read no one does.

      After all, a few "#if defs" here and there is all you need, and to hell with the PS3.5 SP1v2

    9. Re:The answers depend on the questions by n+dot+l · · Score: 1

      The point is you'll be doing work to port the rendering code anyway, so where you start is irrelevant.

    10. Re:The answers depend on the questions by Anonymous Coward · · Score: 0

      Otherwise, you are better off developing in OpenGL, where you can target PCs, PS3, iPhone, iPad, Mac OS X, WebGL, industrial Unix (not all 3D apps are games, dontchaknow?). The only thing you can't do much with is the Xbox (technically possible, but deliberately closed by Microsoft).

      Um, err.. spoken by somebody that sounds like they have never used OpenGL on any of these platforms. But lets make a quick review ok:

      MS-Windows and Linux
          OpenGL4 (AMD 5xxx, NVIDIA 4xx and 5xx cards only)
          OpenGL3 (AMD 3xxx/4xxx, NVIDIA 8/9/1xx/2xx/3xx only)
          OpenGL2 (almost any AMD or NVIDIA you can find now, and most Intel's)

      Mac OS-X
          OpenGL2, that is right no GL3 or GL4 regardless of the hardware inside.
      Some extensions from GL3 make it on some, but most don't. The issue
      is not the hardware but Apple's idiocy.

      Portable:
          OpenGL ES1 (original iPhone, many Android devices, medium range Symbian)
          OpenGL ES2 (iPad, newer iPhone, fancy Android devices, higher end Symbian, N900)

      Browser:
          WebGL: is OpenGL ES2 for JavaScript.

      Here is the thing:
          Each of these is a different API actually. There are some relations:

      GL2.x is almost a strict superset of GLES2
      GL1.4 is almost a strict superset of GLES1
      As a rule of thumb if a device has OpenGL ES2 it also has ES1. ES2 is NOT
      a superset of ES1, but a separate library and header set, where as for desktop
      OpenGL the version is reported. (there is the whole core vs compatibility profile thing
      but lets not get into that!).

      But there are subtle differences for some of the API points that are in both GLES and GL families(even if the function signature is the same, glTexImage family is the most guilty of these).

      Making a desktop GL application and then saying it ports to embedded devices without any work is complete bullshit. Some times the changes are small, sometimes the changes are huge, the biggest pains involve porting OpenGL2 to OpenGL ES2 typically. Also there is plenty of functionality in OpenGL2 that is not in GLES2, and I am not talking just the removal of the fixed function pipeline. Some of this functionality you cannot emulate with other calls (the most guilty being occlusion query).

    11. Re:The answers depend on the questions by ChunderDownunder · · Score: 1
      In so much as the Quartz Compositor on OS X does not support Direct3D, yes.

      But Linux shall. Yes, native Direct3D.

      For those thinking that Direct3D 10/11 on Linux will be sub-par, "Finally, a mature Direct3D 10/11 implementation is intrinsically going to be faster and more reliable than an OpenGL implementation, thanks to the dramatically smaller API and the segregation of all nontrivial work to object creation that the application must perform ahead of time."

    12. Re:The answers depend on the questions by Svartalf · · Score: 1

      As an observation, it's not exactly a good idea to do that- until you remove software patents, etc. from the mix, making a perfect implementation of D3D may trip across patents MS holds.

      --
      I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
    13. Re:The answers depend on the questions by Narishma · · Score: 1

      What do you mean by toolkits? On the PS3 most if not all developers use a low level library called LibGCM. There's an OpenGL ES implementation but it is slower than using the low level libraries so it's only used by a couple very simple games on the PSN.

      --
      Mada mada dane.
    14. Re:The answers depend on the questions by ChunderDownunder · · Score: 1

      Well the big test of software patents ATM is Oracle vs Android. MS will be keenly watching too.

      But I have to ask, how is implementing Direct3D support in Gallium3D any worse than Wine? Or ReactOS? Or Mono? These established projects already re-implement vast portions of Windows. Though RMS does recommend against the use of Mono for that reason.

      I suppose you as an independent game publisher steer clear of those options too...

    15. Re:The answers depend on the questions by Anonymous Coward · · Score: 0

      No, they'll be using LibGCM for everything they can't / don't want to push off to the SPEs. I've heard libgcm described as "like DirectX" too, which is fun.

    16. Re:The answers depend on the questions by SplashMyBandit · · Score: 1

      Your first statement was
      "You can remove the PS3 from that list as nobody uses OpenGL on it."

      Your second statement directly contradicts your first. My point was PS3 either accepts OpenGL or libraries (usually integrated into larger development toolkits) built on OpenGL ES/PSGL. DirectX is not an option for the PS3.

    17. Re:The answers depend on the questions by Narishma · · Score: 1

      My statements do not contradict each other. Developers don't build their games and libraries on OpenGL ES on the PS3, they build them around LibGCM, which is a lower level API than OpenGL. Practically nobody uses OpenGL ES on PS3 as I already said. I don't see what DirectX has to do with anything.

      --
      Mada mada dane.
    18. Re:The answers depend on the questions by Adam+Jorgensen · · Score: 1

      Aha, but doesn't the Wii homebrew development layer use OpenGL?

    19. Re:The answers depend on the questions by grumbel · · Score: 1

      Otherwise, you are better off developing in OpenGL, where you can target PCs, PS3, iPhone, iPad, Mac OS X, WebGL, industrial Unix (not all 3D apps are games, dontchaknow?). The only thing you can't do much with is the Xbox (technically possible, but deliberately closed by Microsoft).

      If you want something truly portable you better have an abstraction layer in between your code and OpenGL so that you don't depend on OpenGL and you can swap it out to DirectX or something different when you like. I mean seriously, a little small homebrew project might only use OpenGL, but everything that is a little bigger will be based on an engine that happens to be more or less library agnostic, so this whole OpenGL vs DirectX talk is really kind of useless. You simply use whatever is the primary API for the platform you develop on.

    20. Re:The answers depend on the questions by Xest · · Score: 1

      "Otherwise, you are better off developing in OpenGL, where you can target PCs, PS3, iPhone, iPad, Mac OS X, WebGL, industrial Unix (not all 3D apps are games, dontchaknow?). The only thing you can't do much with is the Xbox (technically possible, but deliberately closed by Microsoft)."

      This would be true if it weren't for the fact that OpenGL wasn't so painfully fragmented that, well, it's not true.

      Between OpenGL Core profile, OpenGL Compatibility profile, ES 1.x, ES 2.x, Sony and Nintendo's bastardised implementations and all coupled with the fact very useful features often spend too long as extensions all coupled with inconsistency in implementation across hardware means that generally writing OpenGL that spans more than a couple of platforms is no harder than writing DX that spans a couple of platforms.

      I quite like OpenGL, but it's cross platform nature is something that's been repeatedly gang raped into near non-existence over the years so isn't really a valid argument for OpenGL in the first place. You're going to have to abstract a lot of things away if you want to support a reasonable number of platforms regardless, and creating a DX implementation for those abstraction layers is no more difficult than making OpenGL play happily across them.

    21. Re:The answers depend on the questions by SplashMyBandit · · Score: 1

      Your comment is completely valid. However, when I say "cross-platform" I actually don't mean "different versions of Windows and the XBox" (Microsoft's hijacked definition). I actually mean *cross platform*. Yes, true cross-platform isn't ever plain sailing. However, DirectX is simply a non-starter in this area.

      PCs and the Xbox are not the only places to make money (albeit they are big slices of the pie, it is far better to target the *whole pie* instead of of just two big slices). In fact, they are a large but shrinking proportion of general computing out there. So, I use OpenGL because it gives me flexibility, just like it did for Austen Meyer:
      http://techhaze.com/2010/03/interview-with-x-plane-creator-austin-meyer/

    22. Re:The answers depend on the questions by Xest · · Score: 1

      "However, when I say "cross-platform" I actually don't mean "different versions of Windows and the XBox" (Microsoft's hijacked definition). I actually mean *cross platform*."

      Right, so you mean some arbitrary definition that suits your world view then?

      Sorry but just because two platforms are from the same manufacturer doesn't mean it's not cross platform. I completely agree if you don't want to target these platforms that DirectX is a non-starter, but my point is that if you're going to go through the hassle of supporting an OpenGL application across such varied implementation on platforms with such different input methods and so forth then you may as well chuck DX into the fold on Windows/360 anyway.

      Obviously it all depends on context, but when with DirectX you get a unified API for doing full game development then it may well be sensible to just focus on Windows simply because it's the single largest platform anyway.

      Personally I barely use DX or OpenGL nowadays, I'm mostly using XNA because it's such a quick and easy route to developing Windows, Xbox, and WP7- development of games is so quick with it the ratio of speed of development to income as a result of development blows fucking around with DX and OpenGL out the water for indies but different people have different situations.

      What I use DirectX for (which is very little now the work is done) is tools development, because I only care if my tools work in Windows and it gives me the power I need then coupled with MFC it still seems one of the best options for creating tools, although even there I've written a couple of newer tools in XNA now it plays a bit better with the Windows Forms world.

      What I use OpenGL for is Android, but that's been a pain because of the different ES implementations between Android versions, however now everything is moving to 2.0 it's a lot less of a worry.

      Best tool for the job and all that, but if I'm honest I still find working with OpenGL a bit more of a chore than DirectX, but I guess that's really just personal preference, and I'm somewhat spoilt by the pleasure that is XNA.

    23. Re:The answers depend on the questions by SplashMyBandit · · Score: 1

      "However, when I say "cross-platform" I actually don't mean "different versions of Windows and the XBox" (Microsoft's hijacked definition). I actually mean *cross platform*."

      "Right, so you mean some arbitrary definition that suits your world view then?"

      I'm afraid us old-timers laugh derisively when we here stuff like this. No insult intended though. It's just we've been working in this area long enough that Microsoft's re-definition of *cross-platform* is a recent thing (since we've been developing for decades), incidentally their re-definition ('shifting the goal posts, as it were) was intended to produce exactly your reaction (and neuter the arguments of the veterans at that time).

      XNA is indeed nice and OpenGL's lack of support for anything other than graphics and sounds (if we consider OpenAL as well) is pretty lame. It is horses for courses as you say. I'd much rather be able to be productive by re-using all my GLSL skills (which means I don't care about vendor-specific extensions in many cases) all over the place rather than switch tools for each target (which is a style that suits your own workflow, which is cool).

    24. Re:The answers depend on the questions by Xest · · Score: 1

      No need to do the whole condescending "old timers" thing particularly when you have no idea how long I've been in the field myself.

      Historically, a platform has been a much more abstract concept than you're making out, in the academic world phones, consoles, computers have always been called separate platforms because, well, they are - fundamentally the hardware is different, and so is the software, and whilst some components are shared many aren't.

      It's not a re-definition by Microsoft by any measure, it's merely the classic computer science definition of the term. I agree that in some circles that term has been muddied, specifically in that cross platform means different OS, and possibly different architecture but that's a fundamentally narrow and nonsensical definition of cross platform and again, even then with the differences in software between Windows Phone 7, the Xbox, and Windows on the desktop it's not as if it's particularly the same OS anyway. Really, you're just using a bastardised form of the term, when actually all you want to say is that XNA doesn't really work on non-Microsoft platforms, rather than that it's not cross platform, because clearly it is. Personally I prefer to stick to the more rigorous and logically sound terms of academia than misused often illogical slang usually bred from the hobbyist world.

      I'm not switching tools for each target anyway, HLSL/Cg are just fine for the platforms I deal with, it's only really when I focus on Android but as I say that's been very little of my development work thus far. Besides, it's not as if shaders are massive lumps of code that are difficult to port for the most part. It would be really silly to cripple sales to a massive portion of the market over something as trivial as porting shaders.

    25. Re:The answers depend on the questions by SplashMyBandit · · Score: 1

      > It's not a re-definition by Microsoft by any measure, it's merely the classic computer science definition of the term. I agree that in some circles that term has been muddied, specifically in that cross platform means different OS, and possibly different architecture but that's a fundamentally narrow and nonsensical definition of cross platform and again, even then with the differences in software between Windows Phone 7, the Xbox, and Windows on the desktop it's not as if it's particularly the same OS anyway. Really, you're just using a bastardised form of the term, when actually all you want to say is that XNA doesn't really work on non-Microsoft platforms, rather than that it's not cross platform, because clearly it is. Personally I prefer to stick to the more rigorous and logically sound terms of academia than misused often illogical slang usually bred from the hobbyist world.

      Piss off. That's simply bullshit. Sounds until very recently (with phones) you were are a Windows-only developer. Well, I have news for you. In the Enterprise (where the real money is for most companies that aren't Apple, Microsoft, or Sony) it is *cross-platform*. And yes, you sometimes need to deliver 3D and hardware accelerated-graphical algorithms to this world. The bigger the customer is the more diverse their environment is. Old stuff, new stuff, and ancient stuff that is the pillar of their business. You don't get to choose what environment it'll run on. Hence, true cross-platform solutions like OpenGL, Java and Oracle dominate the solutions in this space. The thing is, some of those solutions also work very well on the Windows desktop and phones (eg. OpenGL and Java [called Dalvik for licensing reasons]). Sure, if you are pretty much a Windows-only developer then DirectX and XNA are perfectly fine solutions. But then you have to waste time and *money* re-inventing your solutions when your customers force you to different platforms (Android in your case). IMHO (which clearly is not yours) it is better to start with the cross-platform stuff (which also works perfectly well on Windows if you know what you are doing) and use the time and *money* to do other stuff besides re-inventing the wheel. The IT World is becoming more heterogenous again, not less, so targeting Windows-only solutions makes as much sense as those people targeting "Internet Explorer 6, we don't need W3C" solutions half a decade ago (which seems ridiculous now, yes? yet that is a parallel argument you make for XNA - times are changing yet not everyone is scanning the horizon enough to see it or make sense of what is going on).

    26. Re:The answers depend on the questions by Anonymous Coward · · Score: 0

      Oracle vs Google isn't really about Patents. There are a lot of other grounds Oracle can (and likely will, there's precedent, Sun has sued Microsoft (and won) twice in similar cases involving Java, neither involved patents*)

      i4i vs. Microsoft is the suit you want to be watching, it will decide weather overly general patents can be invalidated (Yes if Microsoft wins, no if they lose). Though I doubt the DX patents are as broad as i4i's "custom XML" patent. And even if Microsoft won that case, you'd have a tough time invalidating the D3D patents, since you'd probably have to use the SGI patents (for IRIS GL, later OpenGL, ironically) which have in all likelihood already been licensed by Microsoft, and don;t mean squat as the patent system allows you to patent improvements to patented tech, provided you have a license for the initial patent.

      Also note that Mono is protected by Microsoft's legally binding Community Promise, all patents required for the implementation of the .NET specification are automatically granted (except to those either suing or being sued by Microsoft, so for example, i4i, barns and noble and foxconn are for the moment, excluded from the MSCP), as well as the patent covenant signed between Microsoft and Novell (who is the main developer of Mono, though I'm uncertain if that deal carries over to Attachmate who has bought Novell).

      Do note that RMS recommends against the use of Mono largely because of the tension between him and Miguel. DO keep in mind that GNU has their own .NET implementation, dotGNU, which RMS does not recommend against using, which is kind of funny, because the GPL3's patent clauses make it incompatible with the MSCP (Microsoft grants you the a patent license, but not permission to grant downstream the same sublicense, downstream is also granted a license from Microsoft, it's purely semantics that comes down to the same same end, but the incompatibility still exists).

      TL;DR version: RMS hates Miguel. Mono is okay, dotGNU isn't.

      * = there was one case between the two involving patents, where Microsoft has to pay out licensing fees on Java patents for .NET. The same VM patents may become a factor here as well, but like I said Dalvik is about more than just patents.

  14. We should not get carried away... by Anonymous Coward · · Score: 0

    ... Because this ain't an apples-for-apples comparison, IMHO nobody got the chance to do a real-world face-off between DX11 and GL4.1 yet. My chips are still all-in on OpenGL, probably because it's cross-platform.

  15. anyone have any actual experience here? by Anonymous Coward · · Score: 1

    just curious, anyone here actually writing code for DirectX or OpenGL? i've written for both and I cannot agree with Mr. Carmack though obviously his knowledge is thousands of times my own. DirectX is an MS API and all that that implies. OpenGL is pretty lightweight by comparison and (like someone else already said) moves a lot of the work into shaders with GLSL where it should be.

    1. Re:anyone have any actual experience here? by Locke2005 · · Score: 1

      Depends on what your criteria for "better" are. DirectX is going to be better able to take advantage of GPU features, so should be faster on newer video cards. OpenGL is going to be more portable.

      --
      I've abandoned my search for truth; now I'm just looking for some useful delusions.
    2. Re:anyone have any actual experience here? by Anonymous Coward · · Score: 0

      OpenGL, with all of its state changes and similar, is no longer a lighter-weight API compared to DirectX.

  16. Because things change by Sycraft-fu · · Score: 5, Informative

    So back in the Doom days, there was no such thing as DirectX. OpenGL was all their was. Of course it was high end cards only, no consumer stuff.

    So move on up to 1996 and the 3dfx Voodoo comes out. It couldn't support full OpenGL, but Glide was based on OpenGL and it brought real 3D to consumers. DirectX was at 3.0 at this point and had no 3D. Glide, or a subset of OpenGL with a wrapper (how Quake did it) was it.

    DirectX 5 came out in mid 1997 and did have 3D, but it was somewhat basic. I mean it could support what consumer cards could do, but lacked a lot that OpenGL had. Still no real comparison.

    However by 2001, DirectX 8 was out and DX was showing some real competition to OpenGL. nVidia had been doing DX and OGL as native APIs for their cards for some time, and both ran just as fast. Also now the cards had programmable vertex and pixel stages, just like the high end pro card, and in fact nVidia was selling their consumer hardware in the pro market as Quadros.

    From there, DX just started pulling further and further ahead. DX10 was a major update and brought some cool new GPU features, like fully unified shaders. Support for it was not a lot on the software side since it required Vista and games have to deal with older computers, but the GPU makers loved it. OpenGL was not fast in terms of catching up.

    DX11 pushes things forward again, and again OpenGL is playing catchup and doing it in a poor fashion with extensions. Not just new graphics features either, but things like support for real multi-threaded and multi-tasking rendering. The ability to treat a GPU much like a CPU and task switch on it and so on.

    Then of course there's DirectCompute, part of DirectX. GPGPU integrated in to the API and the same for all vendors. Of course there is OpenCL, a similar idea, but it is not integrated as DC is in to DX.

    So back when Carmack was an OpenGL fan, it was because it was the best. However it isn't anymore and as things have changed so has his opinions.

    1. Re:Because things change by RyuuzakiTetsuya · · Score: 1

      Not just that but OpenGL wasn't meant as a gaming 3D API, it was meant as an industrial 3D Serious Work API. CAD, CGI, et al.

      When 3dfx came around the corner it was all there was. Khronos has to balance it's interests versus vendors who supply a wide variety of products, not just games.

      Maybe it would be in Khronos' best interest to fork OpenGL into two projects? OpenGL for serious business apps, and some new API for open standards gaming?

      --
      Non impediti ratione cogitationus.
    2. Re:Because things change by dave562 · · Score: 1

      Did you read the article? The gist that I took away from it is that despite the headline claiming that Carmack prefers DX, he does not prefer it enough to redesign the entire codebase to leverage it. Despite all of the improvements in DX, DX is not better enough than OpenGL at this point to make it worth the transition.

    3. Re:Because things change by bhcompy · · Score: 1

      That is why 3dfx made Glide.

    4. Re:Because things change by hedwards · · Score: 1

      It's easy to be the best when you lean on the parties implementing the hardware and writing the software to ignore the other options. Let's be honest about the fact that DirectX won as much because of MS' monopoly position as anything else. Back in the late 90s when this stuff was starting to really kick off there were tons of options available, it's just that MS and DirectX ultimately had the advantage of being able to bundle it with the most popular OS for gaming.

      Most of the options like Glide and PowerSGL were only compatible with a particular brand of hardware and had poor compatibility beyond. In that context, you had basically DirectX and OpenGL available for hardware in a more generic fashion.

    5. Re:Because things change by geekoid · · Score: 1

      This is America, you can't change your opinion just because the data has changed! you keep on making excuses and lie until the gullible people are your supporters.

      Then get a show on Fox.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    6. Re:Because things change by arivanov · · Score: 2

      Maybe it would be in Khronos' best interest to fork OpenGL into two projects? OpenGL for serious business apps, and some new API for open standards gaming?

      No. We are reaching a point where casual software starts relying on 3d accel and hardware video accel. Even browsers and flash (yuck) hook up into accelerated APIs. All of these need the "serious business apps" api, not the "games" api. So regardless of what microsoft thows at DirectX in terms of resources the amount of resources thrown at opengl and opencl in the next five years is likely to grow significantly year on year. Probably even in excess of resources thrown at DirectX.

      So from this perspective, using OpenGL is not such a bad idea for a game manufacturer especially one that is in it for the long run. Add to that mobile gaming that is GL based and has nothing to do with DirectX and the balance of things does not quite look in favour of DirectX if you want to chose today and be in business in 5 years.

      --
      Baker's Law: Misery no longer loves company. Nowadays it insists on it
      http://www.sigsegv.cx/
    7. Re:Because things change by Anonymous Coward · · Score: 0

      ...and again OpenGL is playing catchup and doing it in a poor fashion with extensions.

      *cough* *bullshit*. The big beans for DX11, tessellation, is NOT an extension in GL4, it is
      core and required feature for a GL implementation to report GL4. It is true that GL is playing
      catchup though, but nowadays the GL spec is in people's hands sooner than the hardware
      is usually in consumers hands (but that still sucks since the DX11 jazz was in developers'
      hands much sooner than DX11 hardware existed!).

      Then of course there's DirectCompute, part of DirectX. GPGPU integrated in to the API and the same for all vendors. Of course there is OpenCL, a similar idea, but it is not integrated as DC is in to DX.

      *cough* *more* *bullshit*. OpenCL does integrate into OpenGL: sharing data, sending events, etc. It's called GL/CL interop.

    8. Re:Because things change by mikael · · Score: 1

      Then SGI employed Brian Hook to port OpenGL to PC's to show that a software implementation of OpenGL could be as fast as a software based game engine of the time.

      --
      Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
    9. Re:Because things change by mikael · · Score: 1

      At the started of the 3D graphics chip race, there were 30+ competitiors. 3Dlabs, Nvidia, 3dfx, ATI, Lockheed-Martin, HP, IBM, and many others ... Once the hardware became more complicated and the patents started flying, this was reduced down to a handful. Though, there seems to be resurgence of mobile device manufacturers using OpenGL-ES.

      Microsoft provides far more direct developer support for DirectX along with developer forums and developing console systems. That makes all the difference to developers.

      --
      Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
    10. Re:Because things change by ildon · · Score: 1

      From reading the article, I got that he was acknowledging that it had become better for PC/360, but not better enough for them to bother switching and not better for their design tools anyway.

  17. In other news by Locke2005 · · Score: 4, Funny

    Richard M. Stahlman says vi is now better than emacs.

    --
    I've abandoned my search for truth; now I'm just looking for some useful delusions.
    1. Re:In other news by martin-boundary · · Score: 1

      Neither did Mr Stahlman.

    2. Re:In other news by EnglishTim · · Score: 1

      I can't think of a surer portent of End Times.

  18. This sounds familiar... by Graham+J+-+XVI · · Score: 5, Insightful

    Microsoft had the courage to continue making significant incompatible changes to improve the API, while OpenGL has been held back by compatibility concerns.

    *tweak*

    Apple had the courage to continue making significant incompatible changes to improve OSX, while Windows has been held back by compatibility concerns.

    :)

    1. Re:This sounds familiar... by Robert+Zenz · · Score: 1

      Do they still have the odd Windows 3.11 Font-Installation-Dialog in there?

    2. Re:This sounds familiar... by Ant+P. · · Score: 1

      Supposedly they've actually fixed that in 6.1.

    3. Re:This sounds familiar... by emuls · · Score: 1, Insightful

      And then Windows 7 came out and retained it's compatibility while blowing OSX's socks off. Meanwhile, Apple still only provides software updates for the latest 2 revisions (AKA service packs) of their OS (about 3 years) while microsoft still supports it's 10 year old Windows XP. Hipsters like my former self now regret their mac min and macbook purchases and wish they had just gone with a dell laptop/shuttle pc instead. :)

    4. Re:This sounds familiar... by LordArgon · · Score: 2

      Totally true but, as I understand it, Apple didn't have very lucrative enterprise customers to keep happy. Microsoft's technology has, at times, limited its success, but I believe the converse is also true.

    5. Re:This sounds familiar... by Prune · · Score: 1

      Since around 3.x OpenGL hasn't been held back at all, and any new hardware feature has been immediately made available as an extension by at least NVIDIA's drivers. I know no one that uses only core OpenGL, nor is there any need for it since all compatible hardware supports all the important extensions.

      --
      "Politicians and diapers must be changed often, and for the same reason."
    6. Re:This sounds familiar... by nemesisrocks · · Score: 1

      Apple had the courage to continue making significant incompatible changes to improve OSX, while Windows has been held back by compatibility concerns.

      Have a dose of iOS 4.3, I hope you don't have a pre-3GS iPhone!

    7. Re:This sounds familiar... by Anonymous Coward · · Score: 0

      Except Windows 7 is a better OS than OS X, despite its backwards compatibility legacy.

  19. Re:Actually , the best IDE on the market ... by jjohnson · · Score: 0, Offtopic

    How do you get around, dragging such a huge penis everywhere?

    --
    Anyone who loves or hates any language, platform, or manufacturer, doesn't know what they're talking about.
  20. WebGL, Apple, and Consoles by HRbnjR · · Score: 1

    I don't think OpenGL is going anywhere due to WebGL in the browser, and increasing cross platform development for Apple and console hardware.

    1. Re:WebGL, Apple, and Consoles by Anonymous Coward · · Score: 0

      Nobody said OpenGL is leaving. Btw you are aware that WebGL uses directX on Windows?

  21. Re:Actually , the best IDE on the market ... by Anonymous Coward · · Score: 0

    You too are making the mistake that your chosen toolchain and methods are automatically more productive for everyone else. Therefore, you too should "shut the fuck up". Different people are productive in different ways and through different methods.

  22. Re:Actually , the best IDE on the market ... by Anonymous Coward · · Score: 1

    I don't know about him, but my method is a little wheeled cart that I push in front of me.

    But seriously, all you need is syntax highlighting, searching and a debugger. I've used grep, gedit, and GDB for a long time and they are so quick, I don't have to worry about an IDE that crashes or hangs up randomly.

  23. cross platform by nimbius · · Score: 2

    isnt correct; the article seems to contradict himself. carmack says direct3d is better because of incompatible updates made to the API, where as OP says its multi-platform performance is stellar? let me just load up a copy in my FreeBSD...yeah, that doesnt work.

    his opinion also seems to contradict his own drive toward open source. if the thing you like only works with one vendor, how do you anticipate ever FLoSS'ing your code?

    --
    Good people go to bed earlier.
    1. Re:cross platform by ttyRazor · · Score: 1

      He's talking about OpenGL being good at cross-platform support, not Direct3D. Because Direct3D doesn't have cross platform considerations, that's allowed Microsoft to make more significant changes and pull ahead of OpenGL without worrying about breaking compatibility with other vendors' implementations.

    2. Re:cross platform by Anonymous Coward · · Score: 0

      isnt correct; the article seems to contradict himself. carmack says direct3d is better because of incompatible updates made to the API, where as OP says its multi-platform performance is stellar?

      I see no contradiction. He stated that 'Microsoft had the courage to continue making significant incompatible changes to improve the API,...'. The key part is "improve the API". That means that when MS saw an opportunity to make things better, they did not stop because of the holy goal of eternal compatibility. While I admit compatibility is a great thing if you are running payroll systems, it may not be the most important aspect if you are developing games.

    3. Re:cross platform by Anonymous Coward · · Score: 0

      There's no reason open source code can't call DirectX. Look at SDL

    4. Re:cross platform by Anonymous Coward · · Score: 0

      What a doofus John Carmack is - it's a good job that you were on hand to put him straight.

      He is probably smacking his head with his palm at this very second after being made to look such an idiot by you.
      "No more open source contributions from me .... nimbius made me realise I should just chase Microsoft's money, what with me being so illogical and second-rate, what have I got left, oh god ........"

    5. Re:cross platform by Anonymous Coward · · Score: 0

      his opinion also seems to contradict his own drive toward open source. if the thing you like only works with one vendor, how do you anticipate ever FLoSS'ing your code?

      He seems to be drinking Miguel's Kool-Aid.

      very sad.

    6. Re:cross platform by EnglishTim · · Score: 1

      his opinion also seems to contradict his own drive toward open source. if the thing you like only works with one vendor, how do you anticipate ever FLoSS'ing your code?

      Why would writing his code to target DirectX prevent it from being open sourced?

    7. Re:cross platform by cbhacking · · Score: 1

      I've actually played Direct3D games in FreeBSD (ok, DesktopBSD, which is/was a nice UI over FreeBSD) before. It was through Wine, but it worked just fine.

      --
      There's no place I could be, since I've found Serenity...
    8. Re:cross platform by ConceptJunkie · · Score: 1

      OK, but does SDL call DirectX on Linux? Didn't think so.

      --
      You are in a maze of twisty little passages, all alike.
    9. Re:cross platform by Anonymous Coward · · Score: 0

      isnt correct; the article seems to contradict himself. carmack says direct3d is better because of incompatible updates made to the API, where as OP says its multi-platform performance is stellar?

      No he doesn't.

      From TFA:

      Direct3D handles multi-threading better, and newer versions manage state better.

  24. Haven't read and comments or the story by Anonymous Coward · · Score: 0

    But I'm going to say that Direct3D has been better than OpenGL since Direct3D version 9. I've been waiting for Carmack to say the same.

    And by "better" I mean better to the average gamer and the average developer making games for the most popular platform and OS that is taking advantage of modern GPU features for a modern game.

    OpenGL was left in the dust a long time ago.

  25. JC, MS shill? by TiggertheMad · · Score: 1

    Why is this surprising? He has always struck me as a very smart, pragmatic engineer. If brand x widgets are now better than brand y, any good engineer will switch. His goals seem to be fairly obvious: he wants to build the fastest, most efficient graphic engine he can so he is going to pick the best tools available to do so.

    His honesty is refreshing in a world of shills, ego based flame-wars, and corporate astroturfing.

    --

    HA! I just wasted some of your bandwidth with a frivolous sig!
    1. Re:JC, MS shill? by McTickles · · Score: 0

      Except it wont be fastest graphic engine in Linux because DirectX will have to be translated to OpenGL and that is a major drawback right there, non portability

    2. Re:JC, MS shill? by boristhespider · · Score: 1

      Oh my God! So you mean he'll be missing out on the 0.5% Linux gaming market?? Of which a rough guess of about 90% will happily dual-boot into Windows for gaming purposes? Linux is simply totally insignificant in the gaming industry. Carmac isn't going to develop for Linux if it harms his Windows business (or impedes its progress), because Windows is the platform where something like 95% or more of the desktop gaming market are sitting. The rest of them, on Linux and OSX boxes, can almost all dual-boot and, if they're gamers, probably do.

      What's in it for him to hamstring himself with a set of APIs that are slower than DirectX if not doing so affects less than a percent of his market? He'll go for what's best. He's a businessman, not an OSS fanatic.

    3. Re:JC, MS shill? by McTickles · · Score: 0

      Thats the point right there, he turned from programming guru to businessman ...

      The Linux market is very significant mind you, it is a chicken/egg problem, if there are no games on Linux people will boot in Windows or run Wine, which will make developers think that no one uses Linux when in fact people would gladly play on Linux if it was supported. Any developer with half a brain will understand that.

      In fact it seems that most indy game developers have understood that long ago, and it is not harming them in the least (providing linux binaries), on the contrary it is helping them grow very fast.

      Now the major game ... producing... companies go for familiar and safe, so familiar APIs and the blessings of Microsoft, copy/paste game design, heavy marketing and boom you have another "hit" except not quite so because their new "hit" is the same old thing with more shaders 99% of the time.

      I have a policy myself, any game with a native Linux version I buy, the stuff that runs on Windows on pirate and run in Wine, if it doesn't work in Wine I forget about it all together, my time is too precious to spend playing poorly coded games and I need my money to pay for my own expenses and reward developers that actually make an effort.

      I don't bother dual booting, my computers have better things to do than reboot all the time.

    4. Re:JC, MS shill? by Anonymous Coward · · Score: 0

      your precious pc market is on a short road to the gallows. I'm a counselor, specifically assisting family with develepment challenges for all kinds of kids. Gifted ones, uninspired ones, you name it. People under 20 spend very little time on the PC. I.e. the big box sitting in the room, whirring away. The writing is on the wall. 12 whole years. Windows had a nice little run. Flash in the pan. tablets and smartphones will blow past the global penetration of "The PC", to become the dominant computing platform, and it's already evident, that winders will not be a part of that scene.

    5. Re:JC, MS shill? by boristhespider · · Score: 1

      Hey, it's not my precious PC market, it's just that's the market he's got to aim at. When the PC platform dies he'll be swapping onto Android and the iPhone perfectly happily, because that's the market he'll aim at. I'm not evangelising for any platform -- quite the opposite. And when he does it's likely that, at least at first, OpenGL will be the best tool available to him, so he'll use it again -- the biggest smartphone gaming market is not very likely to be running a version of Windows that can handle Direct3D.....

    6. Re:JC, MS shill? by boristhespider · · Score: 1

      I don't dual-boot much either, but I also don't play many games any more. The ones I do run happily under emulation -- DOSBox and 8 bit and 16 bit computers and consoles, which rather betrays my age. But it's not quite chicken/egg; I've never really bought this "Linux has no games ergo people run Windows" argument. I think the people running Linux will run it anyway, and they'll hit the web with it. When they do their system gets logged... and it's penetration seems to be well under a percent. Spin it however you like, even if *none* of them dual-booted, it's still an insignificant part of the market, and too insignificant to let it derail an attempt to improve the experience for the rest of the market.

      As you say, it's because he's a businessman. I'd say he's still a programming guru, he's just also wanting to make money, and personally I don't blame him. If things change - when Android smartphones become ubiquitous enough to command a good amount of the gaming market, he'll happily swap. He might say, "There were features of Direct3D I wish were in OpenGL," but it won't stop him producing on OpenGL, because it'll be the best tool available to him for a market he thinks significant.

      Indie developers may be doing fine, but generally indie developers are running in a different league entirely. Not meaning to demean them at all, just that it's a different game they're playing. Even the rules are different and only the very basic aim (get people buying my game) is the same. I agree that's how the future gaming scene will look -- multi-platform -- but it'll take a while to get there and it'll be smartphones and browsers driving it more than OSs. Which is a horrible idea, in so many ways...

    7. Re:JC, MS shill? by Anonymous Coward · · Score: 0

      I have a policy myself, any game with a native Linux version I buy, the stuff that runs on Windows on pirate and run in Wine

      So, basically, you're an hypocrite and buys games, no matter how bad they are, just because they run on Linux (I still think you made that part up).

  26. quakeweek predicted this years ago! by Anonymous Coward · · Score: 0

    http://www.quakewiki.net/archives/qweek.planetquake.gamespy.com/034.htm

      Greetings Mr. McGee. I'm Steve Ballmer from Microsoft and I was wondering if id Software would reconsider its stand regarding Direct3D. We have made vast improvements to the API that we think you would appreciate. Do you think you could get Mr. John Carmack on the line?
      tokasd;kjs/;gklsd.msdf;g';asdsd

  27. Re:Actually , the best IDE on the market ... by Zediker · · Score: 2

    this is actually the reason why ye olde time programmers keep away from women and are in such scarce supply. Any erection causes irreversible instant death as all blood is suddenly evacuated from their body into their bulbous appendage. A few die off every year or so, but never as bad as when the internet first caught on, there was a relative extinction event when that occured due to the prevalence of adult entertainment...

    --
    I love to slaughter the english language.
  28. Re:Actually , the best IDE on the market ... by DavidTC · · Score: 1

    If erections last longer than Eternal September, please consult a doctor.

    --
    If corporations are people, aren't stockholders guilty of slavery?
  29. Re:Actually , the best IDE on the market ... by sorak · · Score: 1

    How do you get around, dragging such a huge penis everywhere?

    I have mine carried by a custom-built robot, programmed in assembly, using punch-cards with the holes punched out by the telekinetic power of my...

    Robot: EXTERMINATE! EXTERMINATE, EXTERMINATE!

    ...

    People Screaming: The robot has gone mad, why has this abomination been allowed to live?

    I've got to go...

    People Screaming: Oh the humanity, look what it's doing to the incredibly well-endowed man with the stack of punch cards!

  30. I'm sorry for doom creator by Anonymous Coward · · Score: 0

    I'm sorry for mister doom creator, becouse directx was better platform for windows for a couple of years. If he just noticed then 'hey, man! you're getting old, time to die!'.

  31. directx by Ruede · · Score: 1

    directx is sluggish. i am never seeing a wholesome and clean cycle with direct x games. some where better than others but still... but they are also less (noticeable) jerking when coming under 25 fps carmack is only writing for directx because windows destroyed opengl. that way, he met directx more closely...

  32. Re:Actually , the best IDE on the market ... by Tetsujin · · Score: 1

    How do you get around, dragging such a huge penis everywhere?

    The trick is finding a good holster for your penis.

    It's tougher than you might think - unfortunately your mom's time is spread pretty thin.

    --
    Bow-ties are cool.
  33. Re:Doom Creator? FFS by Provocateur · · Score: 1

    Really. I had to check if it was somebody else other than Carmack!

    You know what you just did, /. ?? From now on, heck, even tonight, a Friday night, no less, he will have to introduce himself in parties as 'Doom Creator' instead of waiting for either the awed silence that usually follows the mention of his name, or the alternative THE CARMACK?!!?

    Did we just fall into Digg, or something...Or did the submitter just want us to *feel* old all of a suddeGET OFF MY LAWN!!

    --
    WARNING: Smartphones have side effects--most of them undocumented.
  34. All Your Base's documentary on id Software. by antdude · · Score: 1
    --
    Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
  35. Re:Actually , the best IDE on the market ... by Anonymous Coward · · Score: 1

    I'm not going to pretend I don't almost exclusively use vim for my development to argue against being stuck in "old ways."
    But there is not a person on the planet who will be able to convince me that typing commands in gdb is faster and more productive than using a visual debugger, even for all its power and scriptability.
    30 years ago the occasional nerd probably argued that they were more productive using ed than that newfangled text editor vi.

  36. Carmack is the best friend OpenGL has ever had by Dr.+Spork · · Score: 3, Interesting

    Geez, if I were an OpenGL developer and Carmack started talking about things that OpenGL should implement to make his game engines work better, I'd be like "Yes sir, Mr. Carmack!" Seriously, those game engines are what's keeping people using OpenGL in the first place. It's too bad that ID software doesn't have the resources to fork that shit and develop it to suit their needs. I'm sure that it would be better.

    It's pretty obvious that the smartest Microsoft engineers are working on game-related projects, and it's smart. Microsoft might be watching its empire erode, but games are a field where their dominance might actually be growing. DirectX is a big part of that, and the Kinect has also really stirred the pot. Lots of comments here are to the effect that Carmack is stating what has been obvious to everyone else for years. Yes, Carmack was a true believer, and his (late) heresy is a sign that MS alternatives in some fields are just ... quixotic. It's not quite like RMS saying that he really should just start using Windows because it works better, but it's about 10% of the way there.

    1. Re:Carmack is the best friend OpenGL has ever had by Daniel+Phillips · · Score: 1

      Microsoft might be watching its empire erode, but games are a field where their dominance might actually be growing.

      Agreed there. In the not too distant future Microsoft's XBox business will be bigger than their OS division, but it won't be because the XBox division grew by orders of magnitude.

      --
      Have you got your LWN subscription yet?
    2. Re:Carmack is the best friend OpenGL has ever had by Daniel+Phillips · · Score: 1

      Carmack was a true believer, and his (late) heresy is a sign that MS alternatives in some fields are just ... quixotic.

      Claiming that alternatives to MS in the game world are quixotic is simply delusional. There is exactly one segment that MS dominates in the game world and that is Windows gaming, the importance of which is steadily declining.

      --
      Have you got your LWN subscription yet?
    3. Re:Carmack is the best friend OpenGL has ever had by drinkypoo · · Score: 1

      Agreed there. In the not too distant future Microsoft's XBox business will be bigger than their OS division, but it won't be because the XBox division grew by orders of magnitude.

      I don't know about orders of magnitude but I'd be shocked if it didn't pick up steam what with Sony doing everything they can to alienate past and potential customers. If you don't want cutesy games then the Wii doesn't cut it. I do want cutesy games, I'm just saying...

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    4. Re:Carmack is the best friend OpenGL has ever had by MikeBabcock · · Score: 1

      And it still won't have paid off the debts incurred by creating it.

      --
      - Michael T. Babcock (Yes, I blog)
    5. Re:Carmack is the best friend OpenGL has ever had by xtracto · · Score: 1

      Carmack was a true believer, and his (late) heresy is a sign that MS alternatives in some fields are just ... quixotic.

      Claiming that alternatives to MS in the game world are quixotic is simply delusional. There is exactly one segment that MS dominates in the game world and that is Windows gaming, the importance of which is steadily declining.

      So, this week PC gaming is declining?
      IIRC last week we argued that PC gaming was increasing and has becamed better than console gaming due to the aging consoles specs.

      And you know... PC gaming == Windows gaming

      --
      Ubuntu is an African word meaning 'I can't configure Debian'
    6. Re:Carmack is the best friend OpenGL has ever had by Daniel+Phillips · · Score: 1

      I don't know about orders of magnitude but I'd be shocked if it didn't pick up steam what with Sony doing everything they can to alienate past and potential customers. If you don't want cutesy games then the Wii doesn't cut it. I do want cutesy games, I'm just saying...

      Microsoft and Sony seem to have teamed up to do the best they can to ensure that this generation is the last hurrah for fat consoles. From here on, fat graphics machines will always be PCs, and later, Arms. Attempting to bring in yet another "supercomputer" console will be financial suicide for whoever tries it. Consoles must be designed as cheap as possible like Nintendo or people won't buy them. Consoles have exactly one thing going for them: convenience. There is no way consoles can ever again compete seriously in the high end gaming arena. Thanks Sony, thanks Microsoft, we don't like the walled garden anyway.

      --
      Have you got your LWN subscription yet?
    7. Re:Carmack is the best friend OpenGL has ever had by Daniel+Phillips · · Score: 1

      Claiming that alternatives to MS in the game world are quixotic is simply delusional. There is exactly one segment that MS dominates in the game world and that is Windows gaming, the importance of which is steadily declining.

      So, this week PC gaming is declining?
      IIRC last week we argued that PC gaming was increasing and has becamed better than console gaming due to the aging consoles specs.

      And you know... PC gaming == Windows gaming

      No, that's as you know. What I know is that Windows gaming has declined in a major way. If you don't believe it then just look at EA's quarterly reports. Consoles passed the Windows gaming platform years ago and Windows gaming has never recovered. Indeed, Windows gaming remains a large but declining market. Future expansion will take place in non-Windows PC gaming platforms. That means Apple and Linux. Yes, Apple runs on PCs now, don't you know? And Linux... think Android on PCs with the latest Radeon and nVidia hardware. It's simply inevitable, and that tidal wave will quickly wash away the dregs of the Windows gaming market.

      --
      Have you got your LWN subscription yet?
    8. Re:Carmack is the best friend OpenGL has ever had by drinkypoo · · Score: 1

      Consoles have exactly one thing going for them: convenience. There is no way consoles can ever again compete seriously in the high end gaming arena.

      I'm not sure I agree with that. I think if they got cheap enough to buy multiples, and if enough games actually supported multi-console multi-screen gaming, then I think people would do that. In fact the consoles could be offered either without an optical drive or in both optical and non-optical models to make it more affordable; non-optical consoles would still be able to download and play games, so some people would never buy anything else.

      Of course, there are major technological hurdles to be jumped, therefore it won't happen any time soon, so I do suspect there will be a lull of the type you describe.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    9. Re:Carmack is the best friend OpenGL has ever had by Daniel+Phillips · · Score: 1

      Consoles have exactly one thing going for them: convenience. There is no way consoles can ever again compete seriously in the high end gaming arena.

      I'm not sure I agree with that. I think if they got cheap enough to buy multiples, and if enough games actually supported multi-console multi-screen gaming, then I think people would do that. In fact the consoles could be offered either without an optical drive or in both optical and non-optical models to make it more affordable; non-optical consoles would still be able to download and play games, so some people would never buy anything else.

      Of course, there are major technological hurdles to be jumped, therefore it won't happen any time soon, so I do suspect there will be a lull of the type you describe.

      That lull is a death sentence. Cheap and powerful are mutually incompatible because of the laws of physics and economics. Of course by powerful I mean in comparison to the PC (and coming soon, Arm) market. It's just not feasible for next generation consoles to compete at the same performance level as PCs. It was barely possible in the current generation and both Microsoft and Sony suffered embarrassing meltdown issues costing billions of dollars. Ballmer would be quietly assassinated by institutional shareholders before they let that happen again, and Sony was so badly wounded that next time round they either show more sense or die.

      Mind, I'm totally happy Sony shot their wad in this generation, it meant I could play games without booting Windows. Next generation is a different story, we don't need Sony any more to get away from Microsoft.

      --
      Have you got your LWN subscription yet?
  37. Irrelevant argument by Kevin+Fishburne · · Score: 1

    This is like arguing about which font is best for printing books. It has absolutely nothing to do with how good a game is, which is what Carmack or anyone in his position should be most concerned about. An extra 10 frames per second or the relative ease of implementing a shader means shit unless as a programmer you're incompetent and/or lazy. If OpenGL or DirectX were used in prolonged scientific simulations for which the efficient use of hardware had a direct impact on time/cost, then it would be meaningful. Add to that the ever-spreading dominance of Android and browser-based games and it begins to sound foolish to use such a restrictive API. When you end up having to port your DirectX code for every project anyway, how do those little advantages stack up then?

    --
    Buy your next Linux PC at eightvirtues.com
    1. Re:Irrelevant argument by McTickles · · Score: 0

      buuuuuttttt, you dont understaaaaand my DX11 shaders (im not goning to mention that we've seen similar shaders in every game before) are better than all the portability in world! *microsoft fanboi whine*

      Game developers have completely lost sight of something very fundamental: is the game fun to play ?

      If you want to make technical demos, join the demo scene (in which using DirectX is seen as a bit inferior)
      If you want to make money, please dont make games, make whatever else, but not games, you are just going to butcher the design in the name of profit.
      If you want to make fun games, DirectX/OpenGL/NCurses/HTML5/Flash it dont fucking matter.
      If you want to make scientific apps, OpenGL and OpenCL, DirectX isn't far from fit for such things.

    2. Re:Irrelevant argument by 91degrees · · Score: 1

      All programmers are lazy. It's almost part of the definition.

      Now, as a programmer, I have a limited amount of time. I can spend that time working out an elegant way to handle the rather clunky extensions mechanism of OpenGL, or I can spend it optimising DirectX.

      If you're porting to a mobile platform, the chances are you'll want to rewrite your entire graphics engine anyway. The architecture is typically somewhat different so the same optimisations may not apply.

    3. Re:Irrelevant argument by Daniel+Phillips · · Score: 1

      Now, as a programmer, I have a limited amount of time. I can spend that time working out an elegant way to handle the rather clunky extensions mechanism of OpenGL, or I can spend it optimising DirectX. If you're porting to a mobile platform, the chances are you'll want to rewrite your entire graphics engine anyway. The architecture is typically somewhat different so the same optimisations may not apply.

      That's an odd tradeoff. You don't want to spend a small amount of time dealing with OpenGL extensions (hint: use Glew) but you are willing to spend a massive amount of time rewriting your game engine? Hint: write it to the EGL spec in the first place then you will also be writing to 3.1 core. Write once, run fast.

      --
      Have you got your LWN subscription yet?
    4. Re:Irrelevant argument by 91degrees · · Score: 1

      You need to rewrite for a mobile platform anyway. Unless you're using pretty basic functionality in which case you might as well use OpenGL in the first place. Except you don't have display lists or immediate rendering. Or fragment shaders (unless you use 2'0 in which case you don't have a matrix stack).

    5. Re:Irrelevant argument by Daniel+Phillips · · Score: 1

      You need to rewrite for a mobile platform anyway. Unless you're using pretty basic functionality in which case you might as well use OpenGL in the first place. Except you don't have display lists or immediate rendering. Or fragment shaders (unless you use 2'0 in which case you don't have a matrix stack).

      If you have to rewrite more than the UI for a mobile platform you made serious design errors in the first place. (And if your game mostly consists of UI then nobody should but it.) Nobody uses display lists or immediate mode for a production game engine any more. Everything is arrays and buffer objects now, which are the same on Opengl ES (sorry for mis-writing EGL earlier) and every version of OpenGL from 2.0 on. Everybody uses their own matrix library now. And soon, every Android phone that matters will support shaders. There are well known techniques for writing lovely 3D graphics without shaders, which is where the first generation touch screen gaming market will sit because it has no choice. Such code works perfectly well on PCs as well, particularly low end netbooks and the like.

      --
      Have you got your LWN subscription yet?
    6. Re:Irrelevant argument by 91degrees · · Score: 1

      And if your game mostly consists of UI then nobody should but it.

      Well, I've worked on one game that fits this description, and was quite successful. (Okay - nothing to do with OpenGL. I'm just nitpicking).

      Everybody uses their own matrix library now.

      Really? A lot of the time you're just drawing objects. When you're doing basic stuff with matrices the fixed function is actually really useful.

      But anyway - while buffer objects are great and all for the main game engine, there's something to be said for immediate mode and/or display lists for some of the less performance dependent stuff. It's pretty handy if you're just drawing boxes and things. Sure it's not that hard to do it with VBOs, but my instinct is to just explicitly set a few vertices, and use a display list if that's too slow. Maybe I'm just showing my age.

    7. Re:Irrelevant argument by Daniel+Phillips · · Score: 1

      while buffer objects are great and all for the main game engine, there's something to be said for immediate mode and/or display lists for some of the less performance dependent stuff.

      Totally agree. I always start by writing stuff in immediate mode, it is significantly less error prone and easier to debug. But I also completely agree that immediate mode should be a separate profile than core, which should always be your production target. If you want some of your core engine to work like immediate mode, then implement a little shell that does it, this is just a quick toss off. Or use one of the many that already exist. When you are ready to productize you can easily justify this finishing work. Being able to have a really clean core tightly coupled to the underlying hardware is a huge blessing.

      --
      Have you got your LWN subscription yet?
  38. Re:Actually , the best IDE on the market ... by bhcompy · · Score: 1

    Look, dipshit, Borland Turbo 4.5 is the best IDE ever.

  39. Other reasons by Dcnjoe60 · · Score: 1

    Is id Software still a player in the game market? I mean there is Doom and Quake, but they are really old. Are they still developing? If not, then why is their opinion on the merits of DirectX or OpenGL that important.

    Personally, I've worked on a few commercial and shareware games that had the choice of DirectX or OpenGL and although DirectX was somewhat easier to code for, the decision was to use OpenGL. My understanding of the issue was that the somewhat increased development time was offset by including the potentially broader non-Microsoft market. While it is true that other operating systems have just a fraction of the market share that Windows has, it is really an untapped market and being a player in that market, with minimal competition was seen as beneficial.

    (Because of NDAs I am not at liberty to state what these games were, so don't even ask).

    1. Re:Other reasons by BitZtream · · Score: 1

      id tech 5 (The doom 4 engine) is used by many games that iD doesn't make. They license it to others much like Epic licenses the Unreal engine to others.

      iD makes the engine, someone else makes the games that use it.

      --
      Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
    2. Re:Other reasons by Anonymous Coward · · Score: 0

      Now, I haven't done any 3d programming since back when OpenGL 2.0 was first coming out... but back then, DEFINTELY OpenGL was easier to program for (state machine and all.)

      I have friends who still do 3d programming, and they tell me that even if OpenGL wasn't required for cross-platform compatibility, they would prefer to use it because it's nicer to work with. JC's opinion here is actually the first I've heard to the contrary from somebody who has (presumably) worked with both. I'm curious to know what the basis is. Also, I've only seen the extensibility of OpenGL referred to as a bad thing before by Microsoft and its fanboys. How is the ability for a graphics card vendor to provide new features without having to wait on a standards body or proprietary library vendor to implement the framework for them "holding back" OpenGL development?

      The only thing I can think of is that maybe if you need to support OpenGL 1.2 in addition to 4.1, maybe you have to know that back then there was no such thing as "ARB_Render_to_Texture" and you have to test for "NV_Render_to_Texture" and "ATI_Render_to_Texture" separately as well. But that's actually easy to solve with a preprocessor macro, and while I haven't checked, I'd actually be slightly surprised if libglu or libglext doesn't provide some sort of automatic backward compatible extension handling like that already.

      (not really surprised, mind you... but a bit)

  40. Been close, at least for a decade now by 91degrees · · Score: 1

    Seriously, once DirectX had support for hardware transform and Lighting it had more or less caught up.

    OpenGL has a slight advantage in ease of setup, and for some applications the fact that it's state based works in the developers favour, but there's not a lot in it. Microsoft has been working very closely with the hardware manufacturers and making sure there's good support for features. The fact that there's a single company with a strong commitment to promoting their API has meant some serious leadership. The OpenGL consortium has been crippled from time to time by various members demanding certain compromises.

    1. Re:Been close, at least for a decade now by Daniel+Phillips · · Score: 1

      The OpenGL consortium has been crippled from time to time by various members demanding certain compromises.

      Khronos group has been moving at breakneck pace (and doing excellent accurate work) for the last few years. That is crippled? I would love to see athletic!

      --
      Have you got your LWN subscription yet?
    2. Re:Been close, at least for a decade now by 91degrees · · Score: 1

      Athletic would have seen most of OpenGL 3 in OpenGL 2. Khronos is playing catchup.

      To be fair, I haven't seen what's in version 4.

    3. Re:Been close, at least for a decade now by Daniel+Phillips · · Score: 1

      Athletic would have seen most of OpenGL 3 in OpenGL 2. Khronos is playing catchup.

      Absolutely true. Before Kronos the history of OpenGL was all about Microsoft bodychecking SGI headfirst into the corner. The landscape is quite different today, and ironically, that is in part due to Microsoft having taken down SGI.

      --
      Have you got your LWN subscription yet?
  41. How terrible by dandart · · Score: 1

    CARRRMAACK!! If you make new games 'doze-only, we'll never forgive you! Because of its terrible lock-in and non-cross-platform, non-vendor-neutral ways, DirectX is not a valid or viable choice for anyone. Last time I looked at photos and specs, GL 4 was better than DX11. I do hope GL looks at itself and makes it more awesome still by sacrificing compatibility a little. But don't break anything. If built with GL3, GL3 it needs to use. Etc. Not that it didn't already... GL4 is awesome. Shame it only works on Fermi GPUs and above. We on 300 or below are stuck with the still-awesome GL3.2. With WebM becoming more popular and DX being locked in to one platform of tens or hundreds now (including consoles, phones etc) and GL working on almost all of them, working with DX now seems stupid.

  42. Profit motive by marco.antonio.costa · · Score: 1

    I guess it's still effective in doing its job.

    I love free software and all the good things it brought us, but I see this happening in lots of projects. It seems like stability and having this 'pure and perfect core' - the thing about the most recent OpenGL and its dependence on extensions for new features comes to mind - is so important that real breakneck-speed, risk-taking innovation doesn't seem to get its turn in open source development, but is rather kept at a distance.

    I guess that has its place, and maybe is the reason that in the server world Linux is so big.

    --
    Send your spendthrift head of state this
  43. Re:Actually , the best IDE on the market ... by MBGMorden · · Score: 2

    Not really. I personally find the autocomplete features in Visual Studio insanely useful.

    Most of my university programming was done in C/C++/Java in a Solaris environment. I run a Linux desktop at home and still do a decent amount of tinkery programming there, and I do maintain some PHP code at work. At home, I tend to use Emacs. At work, on Windows, I tend to use Scite for PHP.

    Those, compared to Visual Studio, are downright painful to code in. If I need to find the name for a random function I have to pull out a quick reference or go to a website. If I want to see all the functions available under an object in C#, I just type the name plus period and then wait a half a second. A little list of all of them pops up next to the cursor. Then once I type the function name, boom - little scrollable list of all the different overload options pops up for that function. All this is done live, which means it even works for some random OSS library I might be used (ie, I've been using MigraDoc and PdfSharp a lot lately, and it's great there).

    Beyond that, the whole RAD environment is just amazing for getting events associate with GUI actions. Doing all that by hand in code isn't that difficult, but man oh man is it TEDIOUS. Why waste the effort when I can double click on a button on a form and it automatically generates a function the runs on the click event?

    For GUI applications, I've simply seen NOTHING that rivals Visual Studio. Now if I'm coding something without a UI (ie, a script that just processes numbers or does backend work), then it's not as valuable, but overall, while I like USING Linux more, I like developing in Windows a lot more.

    --
    "People who think they know everything are very annoying to those of us who do."-Mark Twain
  44. Re:Actually , the best IDE on the market ... by MBGMorden · · Score: 1

    That may have been the worst segue into a "your mom" joke that I've ever seen.

    --
    "People who think they know everything are very annoying to those of us who do."-Mark Twain
  45. Expertise plus open mindedness ... by perpenso · · Score: 1

    In short, "Commander Keen creator says Direct3D is now better than OpenGL" is an argument from authority. Arguments from authority are fallacious since both the reasoning and the premises of the argument may be wrong, and that the argument was made by a famous person doesn't change that.

    What you are missing is that the pro-Direct3D argument being made today is from one of the most enthusiastic proponents of OpenGL of the past, one of the most critical of Direct3D in the past. Also a person of great technical skill in the actual domain being discussed. Its not an argument from authority. Its an argument from experience and expertise, an argument enhanced by open mindedness as suggested by a radical change of opinion as the underlying circumstances changed radically.

    1. Re:Expertise plus open mindedness ... by martin-boundary · · Score: 1
      No, it really is an argument from authority. You have to look at the argument being put forward, not the argument you'd like to see being put forward.

      When somebody says X says Y, that's an argument from authority. The argument is basically the statement "X says Y" here, and this argument doesn't involve the underlying reasons why X might say Y, it just uses the authority of X to claim Y.

      An argument from experience or expertise would have to contain actual statements that describe the experience directly. For example, "X experienced Z,T,U and says Y". You look at the sentence or group of sentences that form the argument, and you see if there's an account of an experience that's relevant to the claim.

      HTH.

    2. Re:Expertise plus open mindedness ... by perpenso · · Score: 1

      An argument from experience or expertise would have to contain actual statements that describe the experience directly. For example, "X experienced Z,T,U and says Y".

      I think that the experience or expertise can be implied for brevity. When Stephen Hawking says "The universe behave thusly .." and a hollywood starlet says "The universe behave thusly .." the arguments are not equivalent. :-)

    3. Re:Expertise plus open mindedness ... by martin-boundary · · Score: 1

      I think that the experience or expertise can be implied for brevity. Stephen Hawking says "The universe behave thusly .." and a hollywood starlet says "The universe behave thusly .." the arguments are not equivalent. :-)

      Implication and brevity is exactly what authority is about. Hawking has authority on the matter, the starlet doesn't. That's why people will quote Hawking in those cases.

      There's nothing wrong with quoting authorities as a shorthand in a conversation, but it's not a valid argument, just an allusion.

      A good way to counter that is to quote some other authority in reply. "Hawking says ... ", then "Ah, but Einstein says ...". Now the first guy has to either attack Einstein, which is a bad move, or he's forced to show his real argument if he has any. Either way you called his bluff and he has to make another move :)

      recommended reading

    4. Re:Expertise plus open mindedness ... by RoFLKOPTr · · Score: 1

      There's nothing wrong with quoting authorities as a shorthand in a conversation, but it's not a valid argument, just an allusion.

      Hey, how about "why the fuck are we arguing over whether somebody's argument about a SUBJECTIVE matter is a fallacy?" He says Direct3D is better, because he thinks it's better. Argument from authority means absolutely nothing here.

  46. What should we be thinking? by erroneus · · Score: 1, Interesting

    Frankly, I don't have the knowledge to comment on the technical aspects of this topic. But as far as philosophical commentary, I might suggest that if there is merit to Carmack's statements, then OpenGL people should sit up and take notice.

    This reminds me of XFree86. Does anyone remember those old farts? For years and years, all we got from them was slow, incremental adjustments and changes. Sure there were improvements but they were so slow in coming and they never seemed to want to commit to anything that looked like a drastic change. Finally, people rebelled and X.org was born and the Linux distros out there haven't looked back. X.org made some great changes from the git-go even if recent improvements seem to have slowed down a bit lately.

    When it comes to display technologies, there is clearly plenty of room for improvement as we go. Frankly, for the desktop, I would rather like to see a departure from the apparently network oriented display model and over in the direction of the hardware driver type model. (Wouldn't it just be awesome if someone could create a Linux display system that could use Windows driver binaries in a wrapper much the way certain network device drivers have been in the past?) I know there has been talk about such graphical interfaces in the past and I am unsure what came of it. I really don't know a lot about these things, but I do know what is lacking and I know that for my purposes, I don't need a networked, multi-user oriented display system on my laptop... just a quick, high-performance display with great 3D graphics.

    So taking into consideration what Carmack says, what could this hint at in terms of what direction Desktop Linux should go? Sure Server Linux should keep X.org and stuff like that -- it has its place without question. And if a new model for graphical display results in being directly compatible with DirectX? Wow, what a neat thing? (I would be afraid of bringing in Windows vulnerabilities and instabilities... would that necessarily be the case? I know with the Windows driver model with things running at ring-0, there is huge potential for big ugly crashes...)

    Anyway...

    1. Re:What should we be thinking? by Prune · · Score: 1

      Frankly, I don't have the knowledge to comment on the technical aspects of this topic.

      You should have stopped right there and not spent the effort writing the rest of the post and boosting FUD. Indeed, you only confirm this later in the post:

      I really don't know a lot about these things

      Then why post? From the point of view of someone that writes both OpenGL and Direct3D, your post makes no valid point, mostly based on a false analogy with X. The problems of X do not map to any corresponding issues with OpenGL. You might as well have used rotten apples for examples.

      And if a new model for graphical display results in being directly compatible with DirectX?
      This is a good example of the nonsense statements in your post. What new model for graphical display??? Any recent version of either OpenGL (3.x/4.x) or Direct3D both are based on the same fundamental architecture! OpenGL brings the programmer just as close to the hardware as Direct3D does. With NVIDIA's direct access extension even the state handling is on the same level, and one can easily create a wrapper to handle that for drivers that don't support it.

      The rest of your last paragraph, in particular, is about as technically sound as Doc Brown's discussion of the flux capacitor in Back to the Future.

      --
      "Politicians and diapers must be changed often, and for the same reason."
    2. Re:What should we be thinking? by erroneus · · Score: 1

      First of all, in X the model I speak of is that command data is transmitted as a stream from process to process either by network link or in memory by sockets. OpenGL is a hardware abstraction layer, but it does not talk to hardware. The "server" in X takes care of that, and whatever hardware capabilities exist get rendered well or not so well at that time. (As I referenced before, X.org has already made serious changes... the "server" is now always the same and a driver is now used for different hardware or even abstractions for display systems such as VNC, NX and more.) But before anything gets displayed on the screen, the program talks through OpenGL which talks to X which determines the session and user and all that mess, which then talks to the hardware specific device driver passing all needed data and then says "okay, draw me a picture that I just described." It's not quite "direct" is it? This is because X is built around the idea of having a remote display device... a terminal... an X terminal. And that model still exists. (The Client-Server model)

      DirectX does something similar, but there is a much closer tie to the hardware as there are memory mapped areas when writing data and graphical commands and such.

      OpenGL does not bring a programmer as close to the hardware as Direct3D. To do so would break the X Window model. If I am wrong, cite where I am wrong. It's not enough to throw about insults.

      OpenGL on Windows is not the same as OpenGL under the X Window environment. So if you are talking about Windows exclusively, we are not talking about the same thing.

      And to be clear, I am talking about "what if there was a way to run native Windows device driver binaries to drive graphics under Linux?" Then the problems of proprietary device drivers not being available to Linux users with equal performance would nearly disappear. But to accomplish that feat, either the X server to driver connection would have to get written, or (because there is a lot of memory mapping going on and expectations that the driver code for devices will run at ring 0) a new graphical display system would have to be developed to replace X.

      Just because you didn't understand what I said doesn't make it nonsense. It just means you didn't understand it. I don't understand French, but it doesn't mean that French is nonsense does it? I get the feeling you are thinking about all of this in Windows. I get that feeling because you seem to think the NVidia driver under Windows is the same as the NVidia driver under Linux. They aren't. They don't perform the same. They don't connect with hardware the same. They don't interface with software the same.

      What are you talking about with this NVidia direct access extension? Do you mean DGA mentioned here http://us.download.nvidia.com/XFree86/Linux-x86/169.04/README/chapter-07.html ? If so, you will see a lot about what doesn't work and why.

      OpenGL is an abstraction API that enables a programmer to write one bunch of code without caring so much about what hardware he's writing to so it could be Nvidia or ATI or Intel or VNC. OpenGL is written expressly for the purpose of removing a programmer from needing to be close to the hardware... or even close to the driver of the hardware.

      So please. If you have something to say, say it -- don't sit back and throw insults without offering proof. It just makes you look like a child.

      As for boosting FUD? Are you talking about the problem of running device drivers at x86's ring 0? That's not FUD, it's fact. The whole purpose of the ring levels is to prevent a program or device driver from taking the whole computer down in the event of some sort of problem. With Windows, a badly behaving driver can affect literally ANY part of the operating system and you wouldn't necessarily know what the cause is. In a properly written OS, it doesn't happen like that -- only the kernel should run at ring 0 whi

    3. Re:What should we be thinking? by Adam+Jorgensen · · Score: 1

      Nice snapback.

      However, some of us actually like the X model. It tends to work better for remote work.

      I guess it's just the sad/happy truth the *nix development does not revolve around keeping game developers happy.

      As for running Windows drivers under a wrapper, why would you want to? You speak like NDISwrapper was worth using.

    4. Re:What should we be thinking? by erroneus · · Score: 1

      The NDISwrapper was worth using as a temporary measure to make use of hardware that would have otherwise been unusable until people were able to figure things out or could finally get assistance from hardware makers.

      Where graphics is concerned, this is a particularly annoying thorn in my side. In my case, "Optimus technology" is under my skin. nVidia "supports Linux users" but managed to slam the door on Linux users faces with Optimus. I can't use my NVidia video on my newest notebook because nVidia won't tell us how to switch to it. Some have learned how to disable the nVidia side so we can at least get by on the Intel video and that's "okay" but not why we bought our high performance machines in the first place. The driver published by nVidia claims to support the hardware in my machine, but it can't "get to it" with Optimus in the way.

      But I wonder -- if we were able to make use of Windows drivers for graphics, there might be a stop gap or even a way of analyzing the processes and changes that the Windows drivers use so that it can be reproduced in some native Linux code.

      As for liking or disliking the X model? I like it too -- it's appropriate for many uses. Desktop doesn't need it through and could benefit with some alternatives.

  47. And irony is .... by ThePhilips · · Score: 1

    Irony is that he said that OpenGL is better - when his opinion mattered.

    It's not that OpenGL now, with the mobile gaming on rise, really cares about what anybody says.

    Or PC gaming, in its perpetual state "it'll get better with the next patch," is now being held back by the choice of 3D API.

    Back then his words mattered. Now they are not.

    --
    All hope abandon ye who enter here.
    1. Re:And irony is .... by Daniel+Phillips · · Score: 1

      Irony is that he said that OpenGL is better - when his opinion mattered.

      Heh, that's it. He is still the goto guy for shadow algorithms but if you want strategic direction on how to succeed in the game world... well, lets just say that Farmville is doing better today than Quake, or even Quake-with-cars.

      --
      Have you got your LWN subscription yet?
    2. Re:And irony is .... by Intrinsic · · Score: 1

      That depends on what your definition of successful is, not everything is about money.

    3. Re:And irony is .... by Daniel+Phillips · · Score: 1

      Well Carmark's id - which forgot how to make games after Carmack pushed out the creative talent one by one - is now owned by Bethesda who rose to dominance on the back of the crappy-but-workable Gamebryo engine. Now what were you saying about successful again?

      --
      Have you got your LWN subscription yet?
  48. OpenGL != open source by Anonymous Coward · · Score: 0

    Just to makes things clear:

    While there is one open source implementation of the OpenGL standard (mesa3d), OpenGL itself does not have anything to do with open source.

    OpenGL is way older then the open source movement and the "open" in its name stands for open standard which is a entirely different story.

    Its funny that some youngsters always seem to jump to the conclusion that anything with "open" in its name must be of course open source.

    1. Re:OpenGL != open source by Anonymous Coward · · Score: 0

      OpenGL was developed in 1992.

      Richard Stallman started the FSF in 1985.

      Donald Knuth released TeX's source before that.

      So... no. The Free Software movement (let alone the 'open source' movement) has been around much longer that you know; if you don't know what you're talking about then keep quiet.

  49. Re:IntelliJ by garethjrowlands · · Score: 1

    I really really like IntelliJ - I'm using it in my current (Java) project all the time. But Visual Studio with Resharper (from the nice people who brought you IntelliJ) has all the good stuff from IntelliJ and that gives Visual Studio a comfortable edge in my opinion.

    IntelliJ does have some useful features I can't seem to find in Visual Studio though - it can visualise WSDL no problem but I can't find the equivalent feature in Visual Studio.

  50. Re:No seriously, shut the fuck up right now. by HornWumpus · · Score: 1

    Code in hex? You pussy. Copy con: to file.exe and code with the alt key and numpad.

    --
    John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
  51. Re:IntelliJ by shutdown+-p+now · · Score: 1

    There are quite a few things in IntelliJ that are not in VS - for example, the awesomeness that is structural code search (it's like language syntax aware regexps, for those not familiar with it).

    But I dare say that IntelliJ mainly focuses on editing code. VS lags behind in that department, but in terms of visual designers it's generally ahead.

  52. Chrome and Firefox would still be popular by judeancodersfront · · Score: 1

    even if they were proprietary. People do not run them because they are open source. The open source religion is really getting old.

    1. Re:Chrome and Firefox would still be popular by MikeBabcock · · Score: 1

      What a fantastic fallacy.

      Chrome and Firefox exist BECAUSE of open source. Were the open source not involved, the products would not exist.

      Android would not exist either, and is out-selling that other proprietary OS.

      --
      - Michael T. Babcock (Yes, I blog)
    2. Re:Chrome and Firefox would still be popular by judeancodersfront · · Score: 1

      That doesn't make any sense. Google could not have built Chrome as a proprietary browser? Why not?

  53. Bullshit by Anonymous Coward · · Score: 0

    I'm an OpenGL developer and cans say without a shadow of a doubt that Carmack is talking absolute shit. In a nutshell Microsoft uses its monopoly position to push DirectX versions. This gives IHV's something to sell and Ctrl-C Ctrl-V magazines something to yap about. But the truth is OpenGL is still ahead (and can use exactly the same functionality via extensions that Microsoft lock out) even if it isn't so well marketed.

    TBH, I'm really surprised Microsoft hasn't been subject to an anti-trust investigation over their treatment of OpenGL. But it's always the same when dealing with bullies. You see the same dynamic with Sarah Palin and Glen Beck. They use their position to control and mouth off to generate the inevitable column inches. So they're "commanding" and "popular"? Not really. Microsoft is the same and Slashdot is just developing Stockholm syndrome with this shill piece. Time to butch up, bitches.

    1. Re:Bullshit by Daniel+Phillips · · Score: 1

      I'm an OpenGL developer and cans say without a shadow of a doubt that Carmack is talking absolute shit.

      Agreed. Why is he doing it? At one time he appeared to have a spine.

      --
      Have you got your LWN subscription yet?
    2. Re:Bullshit by Anonymous Coward · · Score: 0

      Carmack got married and like a lot of sportsmen who get married his performance fell off a cliff. Same deal with Bill Gates. Carmack hasn't made a single notable innovation since then. Megatexture is cool but he's not the only guy to have figured that sort of thing out. Id haven't done a decent game for years and their engine is a load of ass.

      I just think Carmack's brittle and didn't have any staying power. Once he'd sucked his original genius dry he found himself in a vacuum. He stopped learning and his wife provided the emotional prop he used to get from games development. After rationalising this failure he probably figured it was easier to sell out than face his personal reality.

      Tim Sweeny is a twit. George Broussard is roadcrash. Chris Parker is overblown. Shigeru Miyamoto is coasting. All these NT personality types of that generation have hit their limit and exploded. In some cases spectacularly. Their greatest strengths have become their greatest weaknesses. It's happening to Steve Jobs right now.

    3. Re:Bullshit by Daniel+Phillips · · Score: 1

      Megatexture is cool but he's not the only guy to have figured that sort of thing out.

      It's pretty much a direct copy of SGI clip textures. Carmack did make some good progress in shadow mapping, but as you say nothing that others aren't doing at least as well. Carmack continues to excel at optimizing for a relatively static game world with walls for horizons.

      --
      Have you got your LWN subscription yet?
  54. John you are blowing smoke by Daniel+Phillips · · Score: 1

    Sure, he has a point, DirectX may have a temporary lead in multithreaded rendering and state management. But OpenGL is moving up fast. The amount of progress in the last year is mind boggling. OpenGL 4 does a huge amount to close the gap with DirectX, and OpenGL remains far superior in at least one crucial way: cross platform support. Tell me John, how are you going to write Rage for iPhone or Android or PS3 with DirectX? And where do you think the money is in the next generation?

    Also, it is just plain stupid to devalue the importance of backward compatibility. Kronos did the right thing by taking the deprecation/profile path and declining to submit to the imprudent demands of certain loudmouth game coder monkeys. The next battle in the game world will be fought on turf over which Microsoft has no control whatsoever. Anybody who is stuck with a single platform API at that point is going to lose. I have every confidence that OpenGL multicore support will meet or beat DirectX in a short time. And in the mean time, OpenGL is plenty good enough for me, the possibilities are nowhere near exhausted.

    --
    Have you got your LWN subscription yet?
    1. Re:John you are blowing smoke by drinkypoo · · Score: 1

      The next battle in the game world will be fought on turf over which Microsoft has no control whatsoever.

      If you mean mobiles, there's no reason they can't have Direct3D on mobile platforms... in fact, they already do. And since future Windows will run on ARM, it's possible (actually, probable) that future Windows Mobile will be based on the same codebase as real desktop Windows instead of wince. If you mean something else, WTF do you mean?

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    2. Re:John you are blowing smoke by Kaldaien · · Score: 1

      I do not know how you got all that from the article... he clearly states that he has no intention of switching APIs.

      More or less, all he stated was that Direct3D has been driving the evolution of commodity real-time computer graphics hardware for the better part of a decade now. It is a very attractive API if you can limit yourself to a single platform, and deserves respect. This contrasts his (and virtually everyone elses) views on Direct3D from 15 years ago, when the API was a horrible mess. Direct3D was playing catch-up back then, but the situation has been reversed for a very long time.

      The last truly amazing thing I saw (for its time) come out of the OpenGL development pipeline, that could have driven the development of the next generation of GPUs, was ATI's proposal for Uber/superbuffers. Said extension was shot down, and by now the functionality is easily accomplished using a myriad of other extensions. The fact is, Direct3D is the driving force behind what features a new class of hardware will support, and to claim otherwise would be truly foolish. This in no way implies that OpenGL cannot or will not utilize the new functionality, only that Direct3D evolves more quickly.

    3. Re:John you are blowing smoke by Daniel+Phillips · · Score: 1

      I do not know how you got all that from the article... he clearly states that he has no intention of switching APIs.

      I hadn't read the article at the time :-)

      When I did, I realized he isn't stupid enough to switch everything to DirectX, but I also realized he had made a serious PR booboo in letting himself run off at the mouth, likely as not misquoted. There is no way the current state of multicore support and state switching in OpenGL 4 vs DirectX 11 adds up to enough reason to throw away cross platform support. That is no doubt as obvious to John as anybody else. I just can't imagine what got into him to say such a stupid and provocative thing with the certainty of leaving his credibility in tatters.

      --
      Have you got your LWN subscription yet?
  55. APIs are gradually becoming irrelevant anyway... by Kaldaien · · Score: 1

    I completely agree with Carmack, but as Tim Sweeney has stated we are gradually moving to a point where the graphics API is irrelevant. As time goes on, more and more of the GPU is exposed to programability and with each release of DirectX and OpenGL (to a lesser extent), the fixed-function pipeline is shrinking. If the trend continues, we will eventually get to the point where everything is abstracted into programs that run on a GPU and buffers... even state management will become a task managed completely by the engine.

    If you work on cross-platform console engines, chances are you already wrap the state machine into irrelevance to make up for differences between APIs. Multi-threading is simpler in DirectX, but that is to be expected from an API whose design only requires it to work on a single platform. If you code close to the metal, of course you will see benefits for a specific combination of API / hardware.

    OpenGL has traditionally been a monolithic API that tries to maintain legacy support, while accommodating new hardware through extensions. It has a lot of baggage that DirectX does not have, since DirectX routinely drops legacy portions of its API with each major release. Even so, OpenGL comes in multiple flavors now, with the embedded API being much closer to the feature-set modern commodity hardware offers.

    Performance wise, Direct3D has held the crown for many years. Portability wise, OpenGL has always been the king. I do not see this changing anytime soon, the two APIs tackle real-time computer graphics from two very different angles. I, like many engine developers, do not have the luxury of committing to a single API, so it is very difficult to effectively exploit the theoretical advantages of either API.

  56. Re:IntelliJ by garethjrowlands · · Score: 1

    Agreed, there's a lot in IntelliJ but not in VS. But Resharper gives VS most of the IntelliJ goodness. For example, in my copy of Visual Studio structural code search is on the "Resharper/Find/Search with Pattern..." menu.

    So I see VS+Resharper as being ahead of IntelliJ.

  57. World Says CryEngine2 Is Now Better Than idTech5 by G3ckoG33k · · Score: 1

    World Says CryEngine 2 Is Now Better Than idTech 5.

    It looks far better
    It has physics to speak of
    It is more customizable
    It does NOT have megatextures!
    It looks far, far better

    Shit, and I used to like idTech 1,2,3,4, but the Crysis series just blew them out of sight

  58. Valid claim - check out link by Anonymous Coward · · Score: 0

    http://www.incrysis.com/forums/viewtopic.php?id=30645

  59. Re:No seriously, shut the fuck up right now. by petman · · Score: 1

    Code on a PC with keyboard? You pussy. I encode bits directly onto a DVD-R using a handheld laser and a microscope.

  60. Is the API the only thing? by Alarash · · Score: 1

    The API is great, and honestly that's the least you'd expect for something 100% backed by a company such as Microsoft. Not only Direct 3D, but the whole DirectX is successful because Microsoft writes all the functions one needs to develop a game - higher management is very sensible to that kind of perceived money-saving deals. Which is why Windows has been the favored platform for PC games in the last decade. It's a strategy for Microsoft, and it works, fine. The thing is, I'd expect Carmack to be a little bothered by the fact that Direct3D is, obviously, Windows only. I was under the impression that id Software published supported Linux versions of their game not only because they could, but also as a statement. Apparently I was wrong.

  61. API this and Open Source that... by Derpnooner · · Score: 1

    I'm a fan of cross-platform compatibility and I was a huge advocate for OpenGL and open standards - I ditched Windows 98 for a Linux distro when Quake 3 dropped, and I was happy to do so. The statement (late) that John Carmack is making is that DirectX is the refined API to program with, but it doesn't say he has left OpenGL out to pasture (obvious when you see RAGE run on an iPhone).

    I'm all for Open standards, but evolution will always take place. I view this statement as a concern for OpenGL rather than an attack; get the API up to speed and the developers will use it.
    (DOOM 3 / IDTech4 sucked on my Linux distros and Mac OSX) I have faith that the community will abide.

    --
    In Soviet Russia, road forks you!